network system design 2015 mod

*QUESTIONS*

*ANSWERS*

*NOTES*

*NOTES*

*VOCAB*

*VOCAB*

*input/output*

*devices that bring data to and from the CPU (I/O)*

*Which topology would experience much more redundancy in its connections compared to other topologies?*

mesh

*protocol*

method of connection normally Ethernet

*Which topology is better suited for maintaining equal access to resources compared to other topologies?*

ring

*client/server*

server-managed access; the opposite of peer-to-peer

*SMTP*

simple Mail Transmission Protocol; a type of e-mail messaging protocol

*9. needs analysis*

the process of identifying the needs and requirements of a network stakeholder

*10. response time*

the time between entry of a command by a user and delivery of a response to the command

*terabyte*

there are 1,099,511,627,776 bytes in a terabyte

*Most networking components run on how many volts of DC?*

12

*base 3 or ternary*

a number system containing only 3 digits 0, 1 & 2

*4. end user*

a person who uses the network to do actual work

*Which of the following is a definition of a digital phone service?*

a phone service that usually connects directly to a router

*1. closed-ended question*

a question that requires a precise, short, or single-word reply

*Which of the following describes a closed-ended question?*

a question that requires a precise, short, or single-word reply

*subnet mask*

a reverse octet masking system used to limit the number of possible addresses in a TCP/IP network

*Which of the following would most likely use a local area network (LAN)?*

a single high school

*firewall*

a specialized router that allows and disallows traffic; powerful firewalls can also include LAN-wide virus protections, URL filtering, Mal-ware detection, and prevention.

*7. metric*

a standard of measurement

*packet*

a stream of binary data octets; they are also thought of as chunks of data that are most often created by software and not hardware. A packet contains a header which contains logical addressing information, error detection,and correction information. A payload contains the actual data along with TCP and UDP segments.

*bus*

a type of topology characterized by a central backbone to which hosts connect

*motherboard*

the main circuit board in a computer; the board which holds the CPU

*Which of the following would not be a good survey question?*

"Why is network security important to your job function?"

*The Customer Needs Report should contain at least three sections: Introduction, Methodology, and Results. Match the section to the appropriate report content* *1. introduction* *2. methodology* *3. results*

1. a brief description of the project. 2. a description of the people that were interviewed. 3. the Network Overview subsection containing a description of the network as defined by the decision makers

*1. layer 1* *2. layer 3* *3. layer 5* *4. layer 7*

1. the hardware level, electrical specifications, etc. 2. having to do with routing data 3. manages communications between computers 4. services to software when the software utilizes the network

*1. binary* *2. ternary* *3. hexadecimal*

1.a number system containing only 2 digits, 0 and 1 2. a number system containing only 3 digits 0, 1, and 2 3. a number system containing 16 digits 0 - 9 and A - F

How is the base 10 number 32 represented in hexadecimal (base 16)?

20

Octets represent bits of the IP address.

8 eight

*Which of the following is a difference between routers and firewalls?*

A firewall is a protection against unwanted traffic inside of a subnet while a router is not.

Routers & Firewalls In order for a message to reach a recipient on a network, both the sender and receiver must be able to be found in the forest of all users. This is less of an issue on small networks, but the problem becomes exponentially complex on larger LANs and quickly becomes overwhelming between servers and members on the Internet. When the addressing system for ARPAnet (which is the predecessor of our modern Internet) was devised, a saying was coined by Jon Postel in (ARPAnet RFC 791 in September of 1981) A name indicates what we seek. An address indicates where it is. A route indicates how to get there. Further investigations show that the original concept came from another computer scientist, John Shoch, in 1978 from his paper Inter-Network Naming, Addressing, and Routing where he said: The name (what we seek) need not be bound to the address (where it is) until this mapping takes place; the address (or addresses) associated with a particular name may change over time. These are important distinctions to be made. In order for these definitions to hold true, a system of addressing had to be defined with more than one layer. The logical layer applies to the actual machine and can be thought of like a house number. The name applies to the domain which is associated to the logical address by a dynamic system of name association called DNS (dynamic name service). The part about the route in the quote above refers to the way that Internet addressing can be divided into neighborhoods call subnets. Of course, because we are discussing a logical addressing system and not a geographical one, the destinations in an Internet "neighborhood" do not have to exist in close geographical proximity. An IP address contains the information to not only designate a location, it also contains the mathematical indicators with which to determine its location. A router, reads the inception and destination address of each packet it processes and sends that packet along the shortest fastest route possible. A router takes all network pressures into consideration. All traffic on the Internet is managed by a network of routers. The router which stands guard at the entry to a LAN keeps out all traffic that is not specifically addressed to that LAN. A firewall performs a similar service, preventing unauthorized packets from entering a LAN subnet. There are many other sorts of components which you might find on any given network. For the most part, they are either extremely common, like printers, scanners, projectors, data storage devices, etc. They are arcane like medical imaging equipment, automatic teller machines, infrared scanners, motion detectors, a massive array of sensors for vibration, pressure and sound. For the most part, these components are all extensions of a computer or have a computer at their heart. These devices require standard networking controls such as a network interface card, a way to set an IP address, remote access, monitoring, and management.

A network is comprised of physical components, all of which have a purpose or function. It is important for the student to associate the function with the component and to understand the need for the function in the network. Network devices are often plain boxes with no hint of what is going on inside them. In this lesson, we have looked at many of the devices that a network will require and then we have examined the functions of those boxes. Students have been provided history, some background, and some of the underlying theory behind these components. Students are only beginning to know the landscape. As this course progresses, students will be digging deeper and deeper into the functions and the technical theory behind networking and networking tools. You have your feet on the path now. Keep your eyes open for all the signs along the way.

*star*

a type of topology characterized by a central hub and hosts attached each by their own drop or connection

*tree*

a type of topology characterized by a combination of bus and star topologies

*ring*

a type of topology characterized by a ring of hosts

*mesh*

a type of topology characterized by a routed configuration of hosts

*The predecessor of the modern Internet was known as *

ARPAnet

*Which of the following is a difference between computers in the past and computers in the present?*

Addressable memory space has been increasing.

*ARPAnet*

Advanced Research Projects Agency (network)—early predecessor of the Internet

Topologies Topology refers to the configuration of the collection of computers or entities on the network. It is often thought of in terms of shape. The basic topologies are bus, ring, star, tree, and mesh.

Bus Topology This topology uses a central conduit called a bus to connect all hosts on the network. It is characterized by broadcast messages, sent by hosts wanting to communicate. All hosts will see the broadcast, but only the targeted host will respond and process the message. The benefit of this topology is that it only requires a simple cabling scheme. The bus, which is also called the backbone, is the central connector. IT has several drawbacks. IT is only viable up to a limited number of hosts and develops performance issues with numbers greater than twelve. The central conduit or backbone is a single point of failure making it unreliable. Bus topology is used in combination with other network topologies, but it is mostly outmoded.

*Which of the following is an example of computer hardware?*

CPU

*CAN*

Campus Area Network

*CPU*

Central Processing Unit, the calculation processor in a computer

Power As we progress through the components of a computer network, we will find certain concepts emerging and re-emerging from different points of view. Everything in an electronic data processing system requires power. Most all of it is supplied by electricity. Most of the electricity that you will be working with comes from a wall socket. Not all electricity is the same. The power that comes out of your wall socket can be affected by a huge number of influences. Electricity that is free of these influences is referred to as clean (power). Clean refers to power which is relatively free from power spikes, brown outs, 60 cycle hum, extraneous electromagnetic interference, total power line failure, and invasive carrier interferences. Most of these effects are simply the result of power company transformer failures, downed power lines, lightning and other environmental interference, and other problems at the transmission stations or power plants. Some can be caused by malevolent intent. There are differing degrees of clean power, and in fact, it is impossible to eliminate all outside influences on your electricity supply without going to expensive and extreme lengths. For the purposes of this course, and the purposes of most networking environments, a quality uninterruptable power supply (UPS) will provide power which is somewhat clean. Quality UPS devices include lead acid, rechargeable batteries, electrical filtering systems, and solid grounding. Additionally, because networking components are highly susceptible to being momentary power interruptions and most computer devices maintain temporary memory of computations and working data in volatile memory, it is essential that power be constant. In the event that the main power source is shut down, a secure and effective computer network will have a UPS which takes over instantly. As power reserves in the batteries drain, the UPS should report to a log in an attached server or monitoring desktop, eventually initiating an orderly shutdown of the computers attached to the network in the event that the batteries run down before power is restored. The length of time a UPS battery will last is a corollary of size and cost; the bigger the battery, the longer it lasts, the more expensive the UPS. Finally, it is important to note that lead acid batteries have an expected life, but can actually fail at any time. In order to be confident of your UPS system you must run regular tests. More expensive UPS units include software to schedule and perform as well as report on the results of these tests. In the absence of such software, a calendar scheduled manual routine and physical log book will suffice. Also, remember that no system of tests and reporting will do any good if there isn't an attendant process in place to remedy problems as soon as they appear.

Computers It may be hard to imagine, but a computer is really just a calculator. Most of the work that is being done inside a computer is happening within the CPU or central processing unit. All of the rest of the boards, wires, chips, lights, and plugs on the actual desktop box are to facilitate input/output (I/O) functions. Think of it like this. The CPU is a chip of silicone. The chip itself is about one inch square. The rest of the items connected to the CPU chip are there to allow a connection to the hundreds of pins through which the chip communicates with you through the IO devices like the video card and monitor, sound card and speakers, keyboard, mouse and other input devices such as scanners, thumb drives, CD, and DVD drive, etc. The CPU is mounted to a circuit board called the mother board which is also connected to all the other IO devices. The only other part inside your computer is the power supply. Computers run on a stepped down voltage; 12 or 18 volts DC as opposed to 110 volts AC that comes out of the wall socket. In a laptop, the power supply is an adaptor connected to the battery and in a desktop it is an internal transformer connected directly to the wall. Most all networking components run on 12 volts DC. Some have external adaptors like a laptop, but most have internal transformers. Basic troubleshooting technique mandates that when a unit is not working that the technician ascertain the existence of power as a first step. This translates to the old washing machine repairman's adage: "Is it plugged in?" There are so many layers to assure when working with networking components that it is essential to be able to recognize when the power supply has failed. Failed power supplies often, but not always, have a specific smell. Learn what that is and be on the alert for it. Failed power supplies often have a specific unpleasant smell. Be on the alert for it. It is important to note that heat is the enemy of computer network components. Many professional servers include temperature monitoring sensors and can be programmed to report dangerous conditions dangerous conditions via SMS text, pager or email. This information is also available through the server's log files. All events are logged on a server. It is normal and necessary to air condition wiring closets and computer racks.

*What is the difference between HTTP and FTP?*

FTP is for file servers and HTTP is for web servers.

*DAN*

Desk Area Network

*Why is the size of a network not correlated to its cost?*

Different networking technology can be repurposed for wider use.

*DC*

Direct Current; the type of power supplied from a battery (as opposed to Alternating Current or wall socket current)

*DHCP*

Dynamic Host Configuration Protocol (automatic service for assigning IP addresses on a network)

A computer network is not an end unto itself. It is a tool used by real people to accomplish real work. So before rushing headlong into the technical design of a network, it is critical to gather information about who will use the network and how they will use it. From your perspective as a network designer, there are two groups of network users, or stakeholders, who have a special interest in the design of the network. First, there are the decision makers—the company owners and managers who establish the purpose of the network and the ground rules for how it will be created and used. Second are the end users—those people who use the network to do actual work. In a small company, these two groups can, and frequently do, overlap. But even if these two groups are exactly the same people (or person, in the case of a really small company), the needs and concerns of the decision makers are different from those of the end user. Both sets of needs must be considered. In this lesson, you will learn how to conduct a needs analysis in which you gather information about the goals and requirements of the different network stakeholders and use that information to determine what is required in the network design.

Gathering Information It is important to know as much information about the interests, needs, and goals of your stakeholders. While this can be labor-intensive, it will help you in designing a network that will serve your customers well The first step in designing your network is to gather information about the goals and needs of your network stakeholder in as much detail as possible. This can be time-consuming, but you should resist the temptation to skip or short-change this process. What might initially seem like a small straightforward project could easily be much larger than you first envisioned, and adding functionality or physical components after the fact can be expensive at best or impossible at worst. The more information you have before you make any technical design decisions, the more precisely you can define your network requirements, thereby ensuring that nothing is overlooked. You might be able to obtain a small portion of the information you need with a quick Google search on your customer's business name, but for the most part, you will have to use other information-gathering methods such as on-site interviews, focus groups, and written surveys. To be effective, all these methods require basic preplanning to ensure that the information you obtain is both comprehensive and complete. To that end, the remainder of this lesson focuses on how to generate a set of questions to ask the decision makers and end users who are your network stakeholders.

*What should a client understand before setting up his/her network and server architecture?*

His/her software and technology will need to be constantly replaced

Physical network requirements are based on the complex interrelationship between many factors. Business needs of the client organization (number of users, type of work, criticality, scale, or scope of provided services), financial constraint, potential growth, environmental considerations, existing components, existing vs. expanded support, profit model, and work culture/philosophy. Functional requirements can be distilled from the physical needs and can be understood better by considering correlations like the following (Note the underline of the specific network components referenced):

In terms of the user numbers, a small office of five users will need a lower number of servers and switches than a large enterprise employing thousands. Additionally networking across the Wide Area will require specific routing and connectivity that the small office of five would never need to consider. In terms of the type of work, a hospital with its patient data system and imaging storage/transmission needs will require a much more robust network than the hospital parent corporation will need to conduct financial, generic communications, and human resources daily business. There will be vast differences in security provided by subnet separations, routing and firewalls as well as bandwidth between the two even though they are both part of the same business. In terms of criticality, a system which controls the life support in a medical research laboratory for the Center for Disease Control will require far more redundancy (meaning redundant fail over capability for servers, routers, connectivity, power and terminals) and live backup to systems than will a state government driver's licensing website (same functions, smaller scale). But while the medical research lab, which could be responsible for containing Anthrax and Cholera microbes, handles critical life and death services, the state website is just as indispensable to its business model. The cost of losing data (thus requiring solid backups) or loss of the server through attack or failure (thus requiring server redundancy, security, antivirus and monitoring software) would be financially unsustainable. In terms of the scope of provided services, a web server that supports hundreds of fans for a science fiction author will not require the sort of load balancing that a set of collocated web servers will need to support a company like a nationwide company like Groupon or a worldwide one like Ebay/Paypal. In terms of financial constraint, a high-tech service company with twenty salesmen at a call center will need more terminals, servers, printers, backup power, Internet connectivity, and generic productivity software than an auto parts store in a mid-sized American city. But the possibility that the call center could use reconditioned servers and open-source software as well as a shared wireless Internet based fax system could make the individual user's cost much lower for the call center than for the auto parts store, even though the store has a much smaller network with less users. Sometimes size does not correlate directly to cost. Rethinking how to utilize resources In terms of existing components, the entire network could be defined by the single server which was purchased on a five year non-cancelable lease. Server architecture is advancing at an exponentially increasing rate. Newer, cheaper software which requires faster processing and greater memory might not run on an existing server and the company may not have an option of upgrading the server. This constraint can apply to software licensing, printers, Ethernet wiring, etc. Anything with a high capital cost may need to be considered a constant. As a corollary to the point about existing components above, we should not forget the problems created by existing opinions and attitudes. The network engineer must understand that many people in business do not understand or even comprehend network design systems or their interconnected nature. The client mindset is often a challenge to overcome in terms of creating the best solution. In terms of existing versus expanded support, the engineer needs to be aware of the way that new technologies impact old support systems. A company with ten desktop users will be impacted by the implementation of a new CRM (customer resource management) system on many levels besides the cost of customization, implementation, and training. If the company will be internally responsible for support (the opening and tracking of trouble tickets, multiple levels of severity, proper reporting, and re-training as a part of proper issues resolution), the new offering may not help the company at all. The time, personnel, and bottom line cost of these considerations is often overlooked. In terms of the profit model, workplace culture or philosophy, it will impact the decisions the network engineer and designer at a very foundational level. The function of the network must support the physical needs of the employees. These needs will be vastly different in a company that has customers all over the world, which derives its profits from Internet services (which are soft, virtual services often monetized in non-direct ways) and a company with multiple analysts working on a sequential process (like the creation of city government financial reports or digital animation projects or on-line computer game design). In one case, it may be completely acceptable to provide locked down thin-client terminals with no level of customization allowed whereas in another case this might destroy the moral and productivity of the workforce. From this small example, you should be able to see that the subject of network systems design is a complex and interconnected one. Your success in this business and your technical success are intimately tied to your understanding of the functional components of computer networks as well as your understanding of how people will use these tools. And to make it even more complicated, you should have an understanding of user psychology coupled with a healthy dose of business street smarts, because you cannot provide the best solution to a problem unless you know all the factors contributing to the problem in the first place - and many problems are not technical. They are financial and educational. Each Network Appliance has a function and a purpose. Let us go over the basic parts of a network now to make sure we understand what those functions are and what might impact their usefulness. Breaking it down to its most elemental, a network system is the overall structure of connecting computers to each other. Simple, right? But because a computer is somewhat like a human being, we cannot simply connect everything together, if we did that it would be like an individual sharing your eyes, your mouth, your brain, your ears, and even your digestive tract. human body It is a simple fact that though we are trying to connect computers together with a network, we have to be very concerned about what parts we want to keep separated. So let us update the first definition slightly; we will call a computer network system the overall structure shared services and security between connected resources. When we define the computer as a set of shared resources we get a little bit better picture of our actual goal. While it is nice to have complete control of another computer on a network, it is not very helpful to a person trying to use the controlled computer. It is better to think of the connection between shared resources. That implies consent, cooperation, and interconnectivity. Networking is effectively a lot like connecting households together with telephones. Before the advent of telephone switching networks where we all got a personal dedicated phone number, there had to be an operator sitting in a building called a switching station. When you wanted to make a call, you spoke to the operator and she/he would make the connection. This system worked fine until the scale of telephone use grew to where it was no longer feasible to have an individual in between each connection. The system had to be automated in order to grow. A computer network has its roots in the solution to this problem. The protocols which govern the exchange of data between computers works in the same way as an operator did on those elemental telephone communications systems. © 2009 Glynlyon, Inc. Switchboards Using humans as the switchboard operator for computer data would be rather silly, but that's just what we used to do with telephones. When too many folks got telephones though, the process had to be automated. That's how it is now with computer networks. Everything is automated, which is faster and more efficient. The difference is speed. A computer can do its handshake, verification, and error checking with small digital data containers called packets, several thousand times per second. And the speeds are growing every day. The main constraint to faster speeds is the cost of replacing existing systems, not the actual technology for faster communications. This is a very important consideration and one that should always be forefront in the network engineer or network systems designer's mind. It is not the technology with limits us as much as it is out old ways of doing things and the cost of change. Understanding this will make you a much more effective designer. In order to share resources between a collection of computers, you first have to limit the number, type, and location of these computers. For the purposes of example, I will be keeping my description of a computer to include personal computers which is a format which includes servers and desktops, laptops, and mobile devices. Though modern networking still connects mainframe and minicomputers (which are still used in many applications today), I will not be treading into this territory though many of the concepts are the same. There are wide area networks (WAN) and local area networks (LAN). We define local as being within one physical location, often within one IP addressing subnet. We define wide as including many physical locations and multiple subnets. Wide area networks are often thought of as being synonymous with the Internet; local area networks are often thought of as being within one office or your home. There are still networking solutions in use that use early networking connections such Frame Networks. These are dedicated solutions often used within one large company with more than one separate physical location. Frame Networks are beyond the scope of this course except to mention here and remind future network engineers that they still exist and will need to be upgraded when time and money permit. Back to the components and the functions or the different parts of a computer network. Let us talk about some specifics. The method of connection is through a protocol, normally Ethernet these days though there have been others and they may still be used. The term Ethernet has become synonymous with network cable, which is it not. Ethernet is an error-checked protocol which means packets contain a check-sum which is sent along with the packet; if the sending device and receiving device check-sum do not match the packet will be re-sent until they do. The cables used in an Ethernet protocol network must be a six strand, three pair twisted cable called category 5 (referred to in the industry as Cat 5) or higher. Cat 5 cable is traditionally blue. Ethernet is a low-voltage transmission protocol. That means it is affected by other electrical fields (like transformers and florescent light ballasts and the like). Cat 5 cabling is sensitive to sharp corners, knots, and kinks. Ethernet likes to flow smoothly around corners and not be run close to magnets or electromagnetic fields. The buzz that you will sometimes hear on a radio or through a set of headphones is commonly called a 60-cycle hum (meaning 60 cycles per second or 60 Hertz). This sort of disturbance can cause interruptions in Ethernet traffic, causing errors and slowing down network transmissions. Cat 5 cables are either punched down on terminal connection blocks or terminated in plastic quick connectors (the ones you have seen at the end of the cable that plugs the Internet into your laptop) called RJ connectors. RJ stands for Registered Jack. Telephones use an RJ14 and RJ11; Ethernet cables use an RJ45. In addition to the position and the location of the cable in an Ethernet wiring installation, the actual connection to the RJ connector can have a deleterious effect on the data transmission and require inspection. Once we know that our wiring is up to snuff, it is time to inspect what we are connecting to. You cannot simply plug one computer into another through the Ethernet socket and expect communication to occur. At the one end of the system is the demarcation. This is where a WAN would be connected (if your network is to have one). Historically this would be a place in the home office where the telephone company would terminate a copper wire connection device called a smart jack. These were popular when all Internet and WAN connectivity was handled through the analog telephone system built and supported by AT&T. Now a days there are several other options, but originally you either used a modem (which is simply a digital to analog conversion device which translates digital communications of all types into sound and vice versa) or you had one or more dedicated circuits called T1s (or a half Ts), or ISDN (Integrated Digital Services Network). Somewhere around 1995 DSL (Digital Subscriber Line) or ADSL (Asynchronous DSL) and cable modem Internet Service Providers began gaining market share. Any way you look at it, at the WAN connection for your network, you will need a conversion device which takes the raw signal coming in through the wall in a wire and converts it into some sort of usable data transmission line running the Ethernet protocol. Usually this device is thought of as a modem, but it will actually, in most cases, include a router. Packets carry destination and inception address information in the header of the packet (the actual data is referred to as the payload).The router (whether supplied by the service provider or by the subscriber) is essentially a gateway that keeps unauthorized traffic from entering the LAN. The function of a router is to inspect IP addresses and in some cases packet content and route it to its destination by way of the quickest most efficient path. Though it is beyond the scope of this course, there are perimeter routers along a WAN that allow and disallow traffic in the same way. The faster the processor on the router and the more efficient the router tables, the faster the data transmission. So it is necessary that there be a router at the demarcation to your LAN. This protects from unauthorized access (to a point) and more efficiently directs traffic on the LAN. A firewall is just a specialized router. Sometimes firewalls are combined with routers (sometimes they are part of the box we think of as the cable or DSL modem), often they are separate specialized components. Everyone needs a firewall of some sort, the more specialized and complex, the more options you will have. Powerful firewalls can also include LAN-wide virus protections, URL filtering, and Mal-ware detection and prevention. It is necessary for all network traffic to go through the firewall and the router on its way to and from the WAN. In order to accomplish this directional stream of data, it is necessary to put the Firewall and Router in series with one another and to make the LAN side connection to the firewall to be the only connection to the WAN. Though this seems like a simple logical decision, it is astounding to find many installations where someone has circumvented the firewall using a hub or a switch.

Network Management as a Major Component of a Network A network is only as efficient and effective as its monitoring and control. All of the aspects we have spoken about so far would be useless unless the systems of hardware and software, functionality and logic were managed and monitored. Additionally, many businesses make the mistake of monitoring without proper action plans to respond to issues which arise. Though there are no real moving parts in a computer network (aside from on/off switches, disk drive drawers and cooling fans), these systems are subject to degradation due to normal use and due to malevolent attack. Proper management and monitoring are the only way to keep a network running. It is also always better to catch a problem before it affects the whole network. In larger networks, this is partially accomplished by segmentation. As seen by the hierarchal progression of topologies above, the more secure and robust networks are the more segmented networks. It is a physical truth, if you want to limit the spread of a problem or make it easier to find the source, you must divide the whole into compartments. Mesh, tree, and star topologies have natural definitions built into them to accomplish this end. The entire network is connected through the perimeter router (sometimes a simple cable or DSL modem, but a functional router just the same) and very often a server. In very small peer-to-peer networks, the "server" can be one of the hosts on the network. The server and a router are the most common points of monitor and control (which is all the management really is). For example, from the outside of a given network, access in can be accomplished by only two different ways: 1) through a host on the network (through a desktop in the physical location) or 2) through the Internet connection to the router, firewall, DSL, or cable modem. NOTE: It is very important when looking at networks and data processing that we do not miss the forest through the trees. This means, do not be blinded into missing the obvious non-technical solution to what appears to be a high-tech problem. What I am referring to is access. Most small networks have some sort of firewall and authorization. Even a $45 wireless router has an encrypted password. But if you can gain access to the back of that router, you can plug into it and pull an IP address from it (wireless routers distribute IP addresses through a service called DHCP: Dynamic Host Configuration Protocol). Any authorized host on the network can get an address. The router figures that if you can get physical access by plugging in you must be authorized. This may not always be true. Always consider the lock on the door to be the first line of defense. The router and a firewall is a pre-set sort of tool. You define the routing table, you set up what ports you want to leave open, and who you want to be able to access them and the router just does its job. But a router creates a log. And most routers will allow you to set up what is called a syslog (system log) server. This is simply a computer on the network node that is accessible by the router (or whatever device you want to be writing the system log). You establish a routine to have the router write a copy of its log(s) to the server on a periodic basis. Remember: A log is only as helpful as the person who reads it. I can not tell you how many companies create reams and reams of paper, megabytes of log files which nobody ever reads. In order to properly manage, you must monitor. In order to monitor, you have to read the logs. Now there are programs available that will alert you when certain conditions have been met. A free one is called Zone Alarm, it is essentially a software firewall (which is sort of a funny definition since all firewalls are software as are routers, what we buy when we buy a firewall or a router is simply a dedicated computer whose sole purpose is route and inspect packet information. For example, you can set these conditions or alarms to notify you when a certain IP address is attempting and failing to use the proper password. There is a type of attachment called a denial of service attack where the attacking computer is programmed to attempt access repeatedly with random user names and passwords in rapid succession. This can over load the router, firewall, or authentication server and allow a breakdown in security. One way to prevent these attacks is to be alerted when they begin so that counter measures can be taken. There are quiet password breaking schemes as well where an attacker tries a few passwords with a common user name each day. These attacks are often automated as well, so that they come regularly. A monitoring and alert program can flag this sort of issue and send a notification page, email or SMS text to the administrator. This sort of monitoring system is also available as a service online. But, it is important for the network designer to understand the possible threats and formulate the proper procedures to mitigate the risks by managing the network in an orderly and organized way. As said above, not all network problems are malevolent attacks. Many can come through normal software degradation and corruption. Aside from dirty power issues, brownouts and spikes, the most common cause of network degradation comes from the users. It is a network administrator's mantra that a network is designed to help people and people are its greatest enemy. In many cases, you will be required to define systems that protect the users from themselves. Users download most viruses unwittingly. Users overload servers with excess data. Users attempt to configure computer controls in ways that contradict one another. And users, in an attempt to get things done, often do not tell anyone what they have done if they even know themselves. As a corollary to the above, users also will not tolerate being treated as children, bristle at being told no, and will cause problems with no end when you attempt to lock down systems and make it hard for them to personalize their workspace. It is through these non-business personalization's that people cause the biggest problems. With that said, there are certain counter-measures to the user problem that you can design into your networks to lessen headaches down the road. Create solid policies. Window Server versions with Active Directory include very powerful global policy editors. Mandate that certain policies be implemented and be sure that any deviation from these policies is done through a fully documented process with complete documentation of all adjustments. And mandate that these documents are also monitored routinely with a clearly defined action item list when problems are discovered. All Windows servers (and desktops and all Cisco routers and firewalls) have complete and detailed audit trails. You cannot change anything without leaving a thumbprint. These logs should also be monitored regularly. There is no reason that every bit of data on a system cannot be backed up every night, either incrementally or fully every night. These services are cheap and automated and very secure. You can set it up so that all user data is stored only on the server, even if it looks local to the user. You can lock down a server so that no one on the network can see anything that they should not. You can give the server administrator full access. Use strong passwords; make them change every 30 - 45 days. The rule of thumb here is to restrict users as much as they will tolerate (or as much as management will mandate) and after that, protect both the network and the user's data by creating a solid backup of not only the data but also the configuration.

In this lesson, we received an ample overview of different networking architectural models and topographies, as well as, the importance of network management, monitoring, and security (from the design as well as the day to day point of view). We have learned that topology is described by the shape of the nodes or computers in the network and that the shape of the configuration has a logical bearing on the reliability and robustness of the network. This lesson provided an overview of management and security. We will be studying that more in depth as we move forward. The nature of networks is that we can always use the network to maintain the network. It is a self-organizing system with a self-indexing capability as well. So at this point in our study, you should be beginning to see that not only is the network a medium for sharing data but it is also a medium for shared responsibilities. And because the whole concept of networking is an evolutionary process (with fits and starts, detours, mistakes, and corrections), engineers have tried to do it in many different ways. The current methods are, therefore, a summation of all the other attempts that came before. Sometimes this is a good thing, because we have seen where the processes needed improvement and we have had the chance to improve. Sometimes this is an impediment to growth because we have invested so much time and money into doing things one way that changing now would require a revolution. For you as a networking professional, these distinctions and historical details are necessary. Knowing the subject at this granular level will only make you more valuable. Congratulations on coming this far. Get ready for some deeper secrets and details.

LOGICAL NETWORK DESIGN - ADDRESSING AND ROUTING PROTOCOLS To prepare for this class, it would be helpful to think about how numbers and words are merely representatives of physical things. Remember in grammar class how a noun is a person, place, or thing? The word "dog" is a noun. Obviously, if we write the word "dog" it is not a real dog but a representation for that dog. Even a picture of a dog is a representation of the actual animal. How many ways can you represent a dog? Does it belong to a classification? Think about what you know about the classifications in the plant and animal kingdoms. A dog is an animal. It is also a mammal. It belongs to a species, and it also may belong to a particular breed. How does classification change the way you think about a dog?

Internet Protocol (IP) addressing is the core of the modern network. Since its creation in the early days of ARPAnet, it has proven to be a successful and scalable model for creating a routable system for locating resources on the vast computer network we call the Internet. Unfortunately, the creators of IP addressing did not anticipate the enormous speed and popularity of the public Internet because we are running out of IP addresses. In order to understand this problem, you will have to understand how IP addressing works, and we will be taking that subject on in this lesson. You may wonder how it is possible for a well thought out, routable, scalable addressing system like IP to run out of addresses. The short answer is that the scientists that devised it miscalculated the Internet's usefulness. They never thought we would possibly need more than 4,294,967,296 or 232 possible addresses. As hard as it is to believe, we do need more. It is similar to the attitude of IBM scientists in the early days of personal computers who thought that we would never need any more addressable memory that one megabyte. In 2012, the average hard drive in a new computer is 250 megabytes and 2 terabyte external drives are inexpensive and common.The short answer is not hard to understand, but it does not tell the whole story. After we discuss the mathematics behind IP addressing, which is thought of as a logical component of a computer network, we can talk about the solution to this problem. It has been underway for several years. An IP address is a hexadecimal number expressed in four octets. Example: 192.168.1.1 They are called octets because they represent 8 bits of the address. Remember that computers are really just electronic calculators. In any given switch, (the primary state change for electricity) you can only be in one of two positions: on or off. This makes binary (a 2 state counting system, also called base 2) the most natural counting system for a computer. The numbering system that we use for day-to-day math is in base 10. Humans have 10 fingers. It is most natural for humans to use a base 10 system. The base 2 or binary system only has two different digits: zero and one. It is important to understand any value can be represented with only two possible digits. We use the concept of place value to accomplish this. Remember that place value in a base 10 system means that we have 10 possible digits: 0,1,2,3,4,5,6,7,8, and 9. We consider them to be our normal numbers. But to understand the way that computers count, we will have to look deeper than that. We will have to recognize and accept that the Number One represents 1 single item because we have assigned that value to it. In other words, if we were counting transistors in a bowl, the digit 1 represents one single transistor. This may seem obvious and confusing, so let us look at what happens as we add items. See the chart below for clarity. Example 1

As normal as this may seem, it is very important to fully understand that the symbol 10 (which we call ten) is not the only way in which to express a quantity of ten items.

It is not difficult to understand counting to one with the base 2 system. That is the same as in base 10, but as soon as we exhaust the first digit (1) we can only use the zero as a place holder because a zero has no value. That means that the next number in the series, the one that represents two transistors, is 10. It is the same as when we exhaust the digits in base 10. After nine transistors , we have to use the zero as a place holder. Then, the digit 1 can be used to represent ten items by expressing is as 10. Here is one more example to help you get a handle on this concept. Here is one more example to help you get a handle on this concept.

*What is the difference between Layer 4 and Layer 3 of the Open Systems Interconnection layer model?*

Layer 4 communicates data while Layer 3 is routing.

*An area network that spans an entire city is an example of what kind of network?*

MAN

Tree Topology Tree topology uses multiple star configurations connected together on a single bus. In a basic configuration only hub devices (switches & routers) connect to the bus. It is a hybrid networking approach. Its major benefit is greater expandability due to the bus connected devices being at the root of the network. This topology is used in larger situations where expansion will be a key factor.

Mesh Topology This topology uses multiple connections to create a web or mesh of connections. It requires routers at key points in order to create the most efficient pathways to and from each host. A full mesh topology is one where each host is connected to every other host. It is interesting to note that the Internet is a mesh network. Mesh networks are fast and resilient, but they are expensive. Only large enterprises large numbers of host members on their networks require mesh topology.

a standard of measurement is called a *BLANK*

Metric

*MAN*

Metropolitan Area Network

*Why do computers use binary instead of a counting system like base 10?*

Most electrical switches can only have an on and off position.

NETWORK ARCHITECTURE COMPONENTS - PHYSICAL AND FUNCTIONAL One of the perks of working with the hardware of networking is that you get to wear comfortable clothing because you will be physically handling and examining equipment. However, because modern students do not often get physical exercise on a regular basis in order to prepare for a career in networking you might like to consider an exercise and stretching regime to help keep your physical body in good working order and to "wake up" a sleepy mind that needs to keep track of many details. Explore a variety of stretching and cardiovascular exercise regimes.

Network Architecture Components - Physical and Functional. Physical network components refer to the actual pieces of networking equipment like Dell Personal desktop computers, sound cards, video cards, routers, firewalls, switches, cabling, connectors, and network adaptors. In the networking industry, these components are often referred to as hardware. There are many pieces of hardware that comprise a network. Functional network components are what each of these pieces of equipment do. Each component in the long string of computer connectivity has a specific purpose. Understanding these purposes will lead to a more complete picture of what is actually going on in a networked system. When you understand what is going on, it is easier to figure out what is wrong when a system is not working. This sort of overview, "big-picture" understanding also facilitates better design habits (resulting is more effective designs). An effective, efficient design has a greater chance of consistent, dependable operation. In short: a well-designed network yields a happy client. Because computer networks are complex analytical systems, and because electronic data processing has few moving parts, it is difficult to ascertain the purpose or function of a specific network component by looking at it. For the most part, PC computers, routers, and switches are boxes. Some have blinking lights on them that flash occasionally, and cooling fans inside them that hum when they are running, but for the most part, you cannot get much of an idea of what is going on inside the box without understanding the theory behind the component. The purpose of this section is to connect purpose with hardware thus marrying the physical with the functional.

Visit a small office or business that has networked computers. Ask to take a tour. If a network or systems administrator is available and is willing to show you how their systems are set up, this will give you a real world physical example in which to anchor your studies. If not, see if you can guess how their computers work together. Ask questions.

Network Infrastructure To identify your network design requirements, you begin by defining the basic network infrastructure—the underlying hardware and software framework of the network. First you must decide whether your network will follow a peer-to-peer or a client/server network model. A peer-to-peer network requires less hardware and are cheaper to buy and operate, but client/server networks are more stable and provide a higher level of network security and data protection. The decision will be largely decided by the client's needs, which will be defined by using the client survey, an inspection of the existing client's network (if they have one) and the client check list. Aspects to keep in mind can be covered by answering the following questions: What service is the network providing to its users? What is the required speed and bandwidth of the network What special devices are being used, is the current network adequate for the current needs and what additional burdens will the expanded needs put on the current or future network? Finally, it is necessary to determine if the current network conforms to current or future guidelines. Once these aspects are confirmed, you will be able to define the set of hardware and software components that provide both device connectivity and security across the network. Device connectivity Device connectivity refers to the ability of two network entities—devices and/or processes—to make and maintain a connection. Hardware components that provide connectivity are things like Ethernet adaptors, smart jacks, coaxial cable provision points, routers, wireless transmitters and receivers, hubs, switches, cabling, phone lines, power cables, and other hardware that physically transmits data between devices on the network. Network security is provided with either hardware or software components or a combination of the two. Following the requirements provided by the client, a security plan should be defined to determine the level of access and the method of firewall necessary for the client's needs,

For Example A small office running a Windows file and printer server with Internet capability for the five desk top computers and a wireless router for the chief executive will probably not need virtual LANs (VLANS) to keep the internal user data separate from external access because the company is not serving any service to the outside world. Additionally, such a small office is probably not going to set up any complex remote access (programs like GoToMyPc and LogMeIn) can often provide enough security for this sort of office (though these programs are not recommended when any sensitive financial or customer data is stored on desktops or servers.) In a larger, more robust installation, remote access can be managed by a VPN (Virtual Private Network) which provide an encrypted "tunnel" through the public Internet for secure access. In these smaller networks, a simple off the shelf firewall will often suffice and can sometimes be managed by the firewall capabilities provide by the ISP's cable modem or router. Virus detection is typically provided by software components, either separately or as part of the network operating system though in larger installations hardware modules can be implemented to protect entire enterprises from virus, adware and other malware infection. As said, firewalls can be implemented in either hardware or software. To determine what hardware and software components will comprise your network infrastructure, you should consider the following: Size of the network: will there be a need for separate subnets or multiple computer operating systems? Should they be wired, wireless, or a combination? What sort of WAN connection will be required? Does the client have or wish to have standard authentication through a server or will a peer-to-peer situation (with its inherent security risks) suffice? How many hubs and/or switches will you need? Hubs are typically cheaper, but since data throughput timing is always a concern, hubs should be eliminated wherever possible. In a new installation, it is actually best to directly wire each drop (or implement a complete wireless network) where only one high speed (1GB or gigabit) managed switch is required. Managed switches have the benefit of being able to set up isolated software monitoring in order to perform forensic investigations in the case of intrusion; they can often be configured in multiple VLANs; and they can be managed remotely. HP Procurves are used in profusion throughout the industry and are readily available used. What types of transmission protocols must be supported in the network software? If your client is using anything other than TCP/IP (which is the default protocol used on the Internet), make a recommendation to convert his/her network over. It will save multiple headaches and issues in the future. Of course, other protocols might be required for some reason, and cost is always a consideration, but as a network engineer, it is not only your job to look out for the client's best interests and to illustrate the most cost effective solutions, but you also must consider the cost of maintaining the network into the future. That includes your time and the amount of it required to resolve problems. The most common protocol is also the easiest to troubleshoot and expand. Best not to pile multiple layers on top of an already complicated science. What other functions will be supported on the network? If the network will be client/server-based, what type of computer should be used for the central server? Once again, this question will be largely defined by the client surveys, their budget constraints, and what is already in place. But for most small network installations Windows products are common (especially in business applications) though Apple makes the Xserve and Linux, which can provide fast and inexpensive support of any sized network. Deciding how fast should it be and how much memory will it require is often a financial decision in keeping with the more is a better approach. What operating system is eventually chosen is part of a complex negotiation involving money and network management skill (a business which has Widows support capability will be hesitant to install a Linux or Apple server because it cannot be managed by existing resources. The same is true of more arcane operating systems such as Unix or Zenix. This sort of information will be provided by the client as well. The business needs will also define the need for a database or some sort. It is important to understand product offering for cloud based systems, the pros and cons of each, and the subscription costs vs. the capital expenditures required. What type of data protection will be used to ensure the integrity and reliability of network data? Will data transmissions into, out of, and across the network be encrypted? Will data written to digital storage be mirrored? What sort of backups are specified? Should the health of the network be monitored? Would it be better to provide all software as a service via the cloud? Is the provided (or available) bandwidth capable of supporting the business use? The network engineer should be able to make recommendations for third-party solutions to all of these needs. What other functions will be supported on the network? It goes without saying that at this point the engineer would have listed things like shared network printers, centralized digital storage, digital phone service connections (which usually connect directly to the router), or any other device or process that has an impact on the network requirements and performance. Also already covered would be types of generic network support services must be provided? Services such as e-mail, system maintenance services, and task management services (e.g., task scheduling), which are usually provided in the desktop operating system, but not always. Itemize these components to ensure they are not overlooked because they can have an impact on the throughput bandwidth on the network. How much redundancy is required? Which components are most prone to failure and should be considered for duplication on the network? There are other questions you could ask, depending on the size and complexity of the network you are designing. The idea is to generate a complete list of hardware and software components that are required to support the needs of your network users. Specific devices can be listed as well as generic ones. The idea is to get a complete picture of the network's needs - often the client is unaware of what impacts the speed and capacity of the network. It is your job to ask the correct questions and formulate a list of options that provide an array of solutions.

Network Map To understand how all these hardware and software components fit together, you now need to create a map that shows the network organization. On a large sheet of paper, draw a block diagram of all required hardware components. Show the routers that will connect user computers to the network and any hubs and switches that will distribute data. Show any dedicated servers surrounded by those devices they will serve. Include any shared printers, centralized hard drives, VoIP gateways, and specialty hardware devices that will be accessed by multiple users. Mark all computers with any network software components they must have installed. There is no need to make this an artistic masterpiece. Simple boxes work just fine. Include as much detail as you know. Individual user computer computers may be specifically different and as such may require different software, access bandwidth, power, memory, etc. There is no need to use actual networking diagrams, but if you know them it adds a nice touch. Networking professionals use diagraming software to help organize this work. It can be tied to databases in order to keep track of assets (especially in very large enterprise situations). There are some such network diagraming programs available on line. Check them out by searching network diagram drafting or visio type programs. Once you have a complete diagram of all hardware and software components, draw lines connecting those components that interact with each other to show how data will flow between them. Label these lines with the type of data exchanged and the data transmission protocol to be used (if any). Also indicate the type of transmission media that will carry the data, media such as copper wiring, coaxial cable, fiber optics, and wireless communication channels. When you are done, your network map will give you a complete visual representation of all components in your network, as well as the data paths between them.

Bandwidth One of the most important performance metrics of your network design is the set of bandwidths, or data transfer rates, required for transmitting data between the various hardware components. Bandwidth can be really expensive to upgrade after the fact, so you want get it right the first time. Several things go into determining the required transfer rate for a device. The most important factor, of course, is the amount of data that must be transmitted. But bear in mind there is more to this amount than just actual user data. Depending on the device and the data transmission protocol used, data is formatted into packets that contain control information as well as user data. Indeed, in some instances, the control information can amount to more bits of data than the actual user data payload. Plus, some data protocols require multi-pass "handshaking" exchanges for each data transaction. Therefore, it is possible for the number of individual data transactions to have a larger impact on required bandwidth than the actual amount of true data. Required bandwidth is also affected by the method of data delivery. General data services such as routine file management, FTP transfers, and general system maintenance functions do not typically need high bandwidth. On the other hand, interactive applications such as high-performance transaction services (e.g., ATMs or other data retrieval applications), web browsing, and real-time conversational voice and video services all require faster transfer rates to maintain an adequate response time and user satisfaction and to maintain satisfactory image quality. Another factor in determining the required data transfer rate of a device is user competition. Will demand for a particular device be steady throughout the workday, or will it peak at certain times? If the workload is expected to surge periodically throughout the day or week, can users tolerate slower throughput at those times of peak demand? If so, you can specify a lower transfer rate that reflects the device usage during a "normal" work period. But if the increased demand will create an unacceptable bottleneck, you will need to specify a higher bandwidth to adequately cover those times of heavy use, even though that bandwidth will be "wasted" during slower periods. You calculate required bandwidth on a device-by-device basis, and therefore, need to estimate the amount of data to be carried over each data path on your network map. Using what you know of the network users' work patterns, estimate the number of data items that will be transmitted over the path in question during a given time period. Count items such as data files to be saved, documents to be printed, e-mails to be sent—whatever the appropriate data type is for the receiving device on the path. If device usage fluctuates dramatically, you might need to estimate workloads at different times throughout the workday and create an average. Multiply your estimated count(s) by the average size of each item to get the total amount of data to be transmitted. The required bandwidth will reflect that amount of data transferred in the specified amount of time. Note that depending on the type of device, bandwidth might not be specified in terms of bits or bytes. While the data transfer rates of devices such as disk drives, routers, and computer buses are typically expressed in kbits, Mbits, or Gbits per second, not all devices are rated this way. The transfer rate of printers, for example, is usually expressed in pages per minute, and that of scanners can be expressed as inches per second or documents per hour. Express the amount of data that will flow across a particular path in the appropriate measure.

Once you have determined the types of hardware and software components required for your network, you need to compile a Network Design Requirements document that specifies the components required and any performance metrics for those components. The Network Design Requirements document can be as structured or as informal as appropriate for the project, but regardless of its format, the report should include at least the following sections: The Introduction section should contain a brief description of the network. This should include a statement of who the customer is, what they do, how large the company is, and any other general information about the company that pertains to the proposed network. It should also include a brief statement of the primary purpose of the network, including the key business processes that will be supported by the network, as defined by the company decision makers. The Network Infrastructure section should contain a brief description of your network model and framework, including your network map. Explain the relationships between the network components, the various data paths between them, and the data transmission protocols to be used across those paths. The Functional Requirements section should contain an itemized list of all network components, along with any performance requirements for each component. Performance requirements should be precise, unambiguous, realistic, and measurable. That is, do not say "there will be one high-speed printer." Instead, say "there will be one black and white laser printer capable of printing fifty pages a minute." This section should also include a list of any assumptions, constraints, or priorities that might impact the selection of components. Other sections you might add could include a list of what the network is specifically not required to do or a glossary of technical terms. Note that the Network Design Requirements document is not meant to identify specific devices and software packages. Rather, it is intended to itemize the generic types of required network components, describing how they must perform and interact in a way that facilitates selection of those components.

*PAN*

Personal Area Network

Layer 7 Application Layer 6 Presentation Layer 5 Session Layer 4 Transport Layer 3 Network Layer 2 Data Link Layer 1 Physical

Services to software when the software utilizes the network Data representation and formatting code. Manages communications between computers. Reliable transmission of data (disassembly and reassembly of data) Having to do with routing data Links and mechanisms; related to the movement of data as in topology (Ethernet, Token Ring, Thin BNC, etc.) The hardware level, electrical specifications, etc.

What Makes a Good Survey Question Questions? Remind yourself of the best kinds of questions for each situation that will help you do the best job. Before you consider the specific questions you will ask your network stakeholders, you need to understand the two characteristics of a good survey question. First, a good survey question is specific and unambiguous. There should never be any doubt about what you are asking. For example, do not ask an end user, "How many documents do you create?" Instead, ask "How many word processing documents do you create each week?" Second, a good survey question is closed-ended. Unlike an open-ended question that tends to encourage an expansive reply calling on the responder's subjective feelings, a closed-ended question requires a short or single-word reply that precisely answers the question asked. A closed-ended question are definitive and/or quantitative: What will? When will? Which will? How many? How often? How big? As such, they create a set of specific answers that can be cataloged and measured, thereby creating a set of metrics by which your network requirements can be accurately defined. Questions that require a Yes/No or numeric response, or that ask the responder to select from a list of possibilities are good survey questions.

Quick Check: Which of the following are good survey questions? On average, how many e-mails do you send across the network each day? (This is a good question. It is specific, unambiguous and closed-ended.) When do you need this network to be operational? (This is not a good question. It is ambiguous and could be interpreted by the responder in more than one way. A better, more precise question would be "What day do you need this network to be delivered, installed, and operational?") How many of each of the following computing platforms do you use at work? Desktop PC ___ Laptop PC ___ Mac Pro ___ iMac ___ MacBook ___ iPad ___ (This is a good question. It is specific, unambiguous and closed-ended.)What to Ask Decision Makers You begin your information gathering with the decision makers—the company's owners and managers who determine how the network will be used. The answers to the questions you ask the decision makers will define the primary business processes to be supported by the network, as well as any constraints on how the network will be configured, accessed, and used. Bear in mind that in many cases you will not be designing the network from the ground up. You will frequently be working with an existing processing environment, whether networked or not, and that environment will impose certain constraints on your design. Remember to account for this when creating your questions for decision makers, and be sure to include questions about current configurations and procedures.

*RJ connectors*

Registered Jack quick connectors (the ones you have seen at the end of the cable that plugs the Internet into your laptop)

*RDC*

Remote Desktop Connection

*cat 5*

Six-strand, three-pair twisted cable called category 5 used in Ethernet protocol

*open systems interconnection*

interconnectivity model defined in 1947 by the ISO (OSI)

Geographical Today there are LANs, WANs, and MANs (Metropolitan Area Networks). These are all geography-based networks. We have touched on the geographic proximity question with LANs and WANs, but there are several other ways that networks are being considered. Due to the ways people access networks and the advances in hardware speed, storage capacity, and communication speed, it is now possible to think of geography in a more virtual context. In addition to LAN and WAN, we are now considering MAN - Metropolitan Area Networks, SAN - Storage Area Networks, System Area Networks or Server Area Network, (occasionally even considered Small Area Network), CAN - Campus Area Network, Controller Area Network, or even Cluster Area Network, PAN - Personal Area Network, and DAN - Desk Area Network.

Roles Peer-to-peer is also a kind of network. It may seem that roles-based architecture is a simple enough context. After all, different jobs require different resources. And by the same measure, each job can be better facilitated by a different and unique setup of tools. A peer-to-peer network is a loose configuration of communications between computers that are usually in a single LAN, though presumably, you could think of a remote desktop connection (RDC) or a computer to computer connection through the services of GoToMyPC or LogMeIn as peer to peer. In a peer-to-peer network security is specifically managed by each member for that particular member's shared resources. Consistent connection depends upon the different members being available (in that, if your computer is off, the resources you are sharing will not be available to the group). As the name suggests, usually only peers would be connected to one another because executive data should not be available to subordinates and the security restrictions inherent in a peer-to-peer situation are not robust, persistent, or remote manageable. at least not easily, which is why peer-to-peer is not a very good choice in larger, hierarchical, structured environments. A client server arrangement affords all of the connectivity of peer-to-peer with much greater control. The specific controls available in a client server set up are dependent upon the server architecture, but, for example, most all versions going back to MS Windows 2000 server (on the Windows operating system) included individual user authorization, some sort of global security set up in order to restrict or grant access to files and other resources depending upon role and clearance level. Client server includes a huge number of options and customizations. In fact, the biggest single problem with a client server model is the complexity of managing the model and keeping the network under control.

*Which of the following protocols is an e-mail protocol?*

SMTP

A network user who has a special interest, or stake, in how a network is designed is called a *BLANK*

Stakeholder

Ring Topology Ring topology uses a ring of cabling to connect host computers together. Packets progress in the either a clockwise or counterclockwise direction. It is highly unreliable as any cable disruption will affect the entire network. It is very difficult to isolate issue on a ring network. Though some older networks still use this topology (such as FDDI, Token Ring of SONET), ring topography is no longer used.

Star Topology Star topology is the normal configuration for most small offices and home networks. It is the most common topology for wired networks. Though it may use more cabling than bus topography, each host is isolated. Some small office networks break up their star topology into a hybrid bus and star configuration. This should not be confused with tree topology described below. In the small office version, extra hubs and switches are often "stuck" into service rather than placed through planning. The network engineer should be on the lookout for this problem. Usually, these hybrids are created out of necessity in the day to day crunch of running a business. Though quick thinking is to be commended, adding a hub or switch to the end of a star network's host cable creates data bottlenecks and makes network troubleshooting next to impossible. If an extra host is required on a star network, the money must be allotted to add the proper cabling. If the main switch is full, the solution is not to add another switch (unless they are chained together at the gigabit level). Hubs, switches and cabling affect all areas of network performance and must not be left to chance.

*SAN*

Storage Area Network

Though they are often built into the motherboard in newer computers, the network adaptor is the connection point between the individual computer and the network at large. The card will determine the protocols that can be used, and the speed of transmission. In some cases, a network adaptor can be used as a bootable device in order to start a computer without accessing its hard drive and installed operating system. While this ability can facilitate the removal configuration of computers in a large office or an enterprise wide network, it also opens up a weakness in desktop computer security that must be considered in the big picture. A network adaptor is such a common component in all computers that the end user is often not aware that it exists. The other component which serves the same purpose is a wireless adaptor. It is important to note that due to the popularity of USB connectivity, all adaptors and almost any input output device can be connected through the USB port. USB stands for universal service bus.

Switches and Hubs A hub basically operates on the same sort of idea as a splitter on a cable TV line or a multiple plug extension cord for general electrical power connections. The hub is doing a little bit more however because packet transmission contains two way communications. Hubs usually have an external power adaptor and often have blinking lights that are supposed to be showing data transmission rates and packet collisions. The bottom line to remember about a hub is that it contains no internal logic systems. It will do mothering to balance the transactions following through it. That means that if you are supporting 10 Mbps on one Ethernet Cat5 cable and you split it into connection for three computers, the hub will only split the supported throughput into thirds. This could have a huge negative impact on the devices connected to the hub. Even though they are an inexpensive way to add devices to your network without additional wiring costs, professionals recommend that it would be best to avoid hubs altogether and replace them whenever possible. Switches, on the other hand, are computing devices. They perform a degree of load balancing, meaning that when there is a call for more bandwidth from a specific device plugged into it, the switch does this sort of minor routing automatically. Depending upon the quality of the switch and the features it may have been designed with, you may also create more detailed rules to govern this minor routing. Additionally, a managed switch (like the HP Procurve line or the Cisco Stackable Managed series) will include VLAN capability, secure access, remote management, internal port monitoring, and interconnected redundancy. It is important to understand that switches are a smart technology (in that they do a certain amount of signal processing) while hubs are a dumb technology (because they just split signals without any consideration for the data being carried). It is always better to use a smart solution.

7 Application Layer 6 Presentation Layer 5 Session Layer 4 Transport Layer 3 Network Layer 2 Data Link Layer 1 Physical Layer

The OSI (Open Systems Interconnection) layer model was defined by the International Organization for Standards (ISO) in 1947. It is used to define different modes of interconnection between components in a networking system. It is a system independent of the protocols and applications running on it. Though we will only be discussing TCP/IP, the OSI model governs any sort of addressing or protocol and assures interconnectivity between a multitude of different networks. Here is a very brief description of the different layers. A clever saying or mnemonic to help remember these layer names is All People Seem To Need Data Processing Layer 7 Application Services to software when the software utilizes the network Layer 6 Presentation Data representation and formatting code. Layer 5 Session Manages communications between computers. Layer 4 Transport Reliable transmission of data (disassembly and reassembly of data) Layer 3 Network Having to do with routing data Layer 2 Data Link Links and mechanisms; related to the movement of data as in topology (Ethernet, Token Ring, Thin BNC, etc.) Layer 1 Physical The hardware level, electrical specifications, etc. The network is the only layer concerned with IP addressing. But because TCP (Transmission Control Protocol) is integrally connected to the process of IP address use (Thus the designations TCP/IP) and the session and transport layers concern themselves with TCP. We will work more with this model going forward. For now, we only need be concerned with the way IP addresses interact with devices at these layers. The layer 3 Network layer is all about delivery. It provides information about the particular protocol used, and local data integrity ensuring a checksum. Internet Protocol (IP) is a Network layer protocol. Routers operate at this layer by inspecting addressing information before making a decision on where a packet should be forwarded. Layer 3 switches also base forwarding decisions on this information. Layer 3 routers and switches ignore HTTP, FTP, or SMTP data. A router (or layer 3 switch) only cares about where the packet is coming from and going to. Classful Addressing and Subnetting, Routing is a major functional component of a network. It accomplishes this by reading individual packets of data, inspecting them for their inception and destination, and forwarding them on by way of the most efficient route. The IP address is a self-directing address scheme. IP addresses are the element which tells the router what to do with the datagram of the packet. One way that routers determine information destination is by class. Classful IP addressing is the division of an address space (an address space is defined using a subnet mask: (0.0.0.0 - 255.255.255.255) into contiguous ranges, also thought of as classes. Contiguous refers to the lack of gaps or missing addresses between the first and last address in the range. Using this distinction it is possible to identify the network portion of the address by inspecting the first 4 bits of the first octet. These first 4 bits are referred to as the "most significant bits," and define the class of the address. The value of the first 4 bits determines the range for the specific class of the address. Using this information, a router can determine the part of the address that identifies the network and the part of the address that identifies the actual host. Subnetting is a method of limiting the definition of the range that a particular network occupies. For example, if your LAN used a complete range of addresses from 192.168.1 to 192.168.1.255 (note that the first three octets do not change) you would have a total of 254 addresses to work with. The subnet mask is a mechanism to limit the number of possible addresses the router considers in any particular network. It is dependent upon the gateway address in order to work. The gateway tells the router what the first address in the range is. The mask tells the router, by default, how many addresses will be in the range. If all the addresses noted above as being (or may possibly be) used, then the subnet mask will be 255.255.255.0 . The term mask refers to the fact that it is reverse numbering scheme. Since there are a possible 255 different digits in each octet of an IP address, then a mask with 255 in the first octet says that this address will never change; there is no need to look at it in any inspected packet if it does not match the first octet of the gateway. So forth and so on down each octet until the last one.

*electromagnetic interference*

interference from motors and other electrical magnetic oscillations

*60 cycle hum*

interference that effects the smooth operation of an electrical circuit

The quantity in the bowl One transistor Two transistors Three transistors Four transistors Five transistors Six transistors Seven transistors Eight transistors Nine transistors Ten transistors Eleven transistors base 16 base 16 Two transistors Three transistors Four transistors Five transistors Six transistors Seven transistors Eight transistors Nine transistors Ten transistors Eleven transistors Twelve transistors Thirteen transistors Fourteen transistors Fifteen transistors Sixteen transistors Seventeen transistors base 16 base 16 Two transistors Three transistors Four transistors Five transistors Six transistors Seven transistors Eight transistors Nine transistors Ten transistors Eleven transistors Twelve transistors Thirteen transistors Fourteen transistors Fifteen transistors Sixteen transistors Seventeen transistors base 16 base 16 Two transistors Three transistors Four transistors Five transistors Six transistors Seven transistors Eight transistors Nine transistors Ten transistors Eleven transistors Twelve transistors Thirteen transistors Fourteen transistors Fifteen transistors Sixteen transistors Seventeen transistors

The number used to represent that quantity in base 10 1 2 3 4 5 6 7 8 9 10(here is where place value kicks in) 11 (this is just to continue the pattern for comparison) base 2 1 10 11 100 101 110 111 1000 1001 1010 1011 1100 1101 1110 1111 10000 10001 base 3 1 2 2 11 12 20 21 22 100 101 102 110 111 112 120 121 122 base 16 1 2 3 4 5 6 7 8 9 A B C D E F 10 11

*In a system that has been maintained and protected, what should happen when the system goes down?*

The server redundancy will run its backup.

*VLAN*

Virtual LAN; logical separate networks usually configured with in a router, firewall, or switch for security purposes

*A network map shows the network organization. Which of the following would not be included in a network map?*

malware

NETWORK ARCHITECTURAL MODELS - TOPOLOGIES AND CLASSIFICATIONS Review what you already know about Network Design. Examine any Network Requirement documents that you might have available to you. If none are available, create a fictitious network and think about the various roles and needs of the various participants in that network including human, hardware, and software.

There are three major network architectural models. Those based on geographic proximity (LAN, WAN), those based on host's roles (peer‐to‐peer, client/server), and models based on network topologies.

*The Functional Requirements section of the Network Design Requirements document contains a list of the network components and their performance requirements. Which of the following would be a good example of a performance requirement?*

There will be one color laser printer capable of printing twenty-five pages per minute.

Customer Needs Report Once you have formulated the questions you want to ask, interview your network stakeholders or have them complete a written survey. Then collect and quantify the answers, summarizing them in a Customer Needs Report. The purpose of this report is two-fold: It describes exactly what the network stakeholders expect and need from the network and serves as the basis for a formal agreement between you and the customer's decision makers. It provides the necessary parameters for determining network performance requirements that will drive the ultimate technical network design. (Determining network performance requirements is covered in the Network Design Requirements lesson.) The Customer Needs Report can be as structured or as informal as appropriate for the project, but regardless of its format, the report should include the following sections: The Introduction section should contain a brief description of the project. This should include a statement of who the customer is, what they do, how large the company is, and any other general information about the company that pertains to the proposed network. It should also include a brief statement of the primary purpose of the network, including the key business processes that will be supported by the network, as defined by the company decision makers. The Methodology section should contain a description of how you conducted your needs survey. How many decision makers did you interview? Who were they and what role do they fill in the company's organizational structure? What questions did you ask? If the number of questions is very large, you might want to simply list the types of questions here and append the actual list to the end of the report. How many end users did you interview? Break the number down by end user type—application users, developers, administrators, etc. What tasks did you identify for each type of end user? What questions did you ask about each task? Again, if the number of questions is very large, you might want to simply summarize the questions for each task and append the complete list to the end of the report. Results The Results section should be divided into two subsections: The Network Overview subsection should contain a description of the network as defined by the decision makers. This should include a restatement of the network purpose (as given in the Introduction), as well as descriptions of any conditions and constraints the decision makers have put on the network design. Quantify these conditions and constraints, wherever possible. If you interviewed more than one decision maker, summarize their responses into a single brief statement for each question you asked. The User Needs subsection should contain a description of what the network end users will need, grouped according to each task performed. Quantify and combine the responses you got from the end users you interviewed to present a detailed picture of what those users will require to complete their tasks. Break this information down into as many discrete metrics as you can think of: How many e-mails a day will be sent across the network? How much data will be sent to network printers during the course of a typical day? How many users will be logged onto the network, broken down by time of day? How you organize this section will depend on the network size and complexity, but you should present the information in a way that gives you a detailed picture of user requirements. Note that the Customer Needs Report is merely a summary of what the network stakeholders have told you they need the network to do. It does not address any performance or technical design issues, but lays the groundwork for determining those metrics, as covered in the Network Design Requirements lesson.

To design a successful network, you must first understand precisely what the network will be used for and what the needs are of those people who will use it. In this lesson you have learned: who the network stakeholders are that you need to interview; how to craft a good interview question; what types of questions to ask of a network decision maker; what types of questions to ask of a network end user; and how to write a Customer Needs Report that summarizes what the network stakeholders told you they need from the network.

*Response time is the time between entry of a command by a user and delivery of a response to the command.*

True

*UPS*

Uninterruptable Power Supply; also known as a battery backup often containing computerized circuits to optimize the transfer to and from battery power in the event of a power outage; also contains power filtering circuits to assure clean power

*router*

a networking device that sorts packets of data by way of their inception and destination addresses

*base 16 or hexadecimal*

a number system containing 16 digits 0 - 9 & A - F

When computers were first considered, scientists recognized that although an electrical switch can only have two positions, we would need slightly larger "words" in order to create a computer language. It was decided to establish the smallest computer word as one byte which is equal to eight bits. A bit is the representation of a single switch in a computational electronic circuit. Thus, we created machines that speak in binary words that are eight bits long (binary which means 2 X 8 bits= 16 possible values that can be expressed in one byte, the basic computer word.) It makes more sense for the numbering systems for computers to be in base 16 or hexadecimal. The big difference between base 16 and base 10 is that we only have 9 possible digits (plus the place marker zero) to represent the values from 1 - 10. In hexadecimal we need 15. They are only representational - we could use almost anything (®, §, £ or ,, and). But the most universal extension of the Arabic number system was chosen to be the alphabetical letters A, B, C, D, E, and F. Remember that these are representations of quantities. It bends the brain to see letters in a mathematical equation at first. See the comparison below.

We are currently using IPV4 (Internet Protocol version 4) which has 232 possible values. There is a transition going on to change over t IPV6 which will have 2128 possible values in sixteen octets. That translates into 3.403×1038 unique addresses or 340,282,366,920,938,463,463,374,607,431,768,211,456 - it should be a while before we need any more than that, but you can never tell. Obviously, the IP address is a very important functional component in a network. Without it, each individual computer would not be able to be located. What's more, without an addressing scheme like IP, a data request on the network would not be able to locate the requestor: the data would never be able to be returned. That is the basic idea behind IP addressing. Let us address how it is done. An Introduction to The Seven Layer OSI Model

*Which of the following concerns of decision makers would prompt the inclusion of scalability questions in a needs analysis survey?*

What future planning issues should be considered?

*Which of the following questions is the most important to consider when deciding between a peer-to-peer or client/server network?*

What is the required speed and bandwidth of the network?

Additional Things to Consider are the Following... How will the network be kept secure? VO: How vigorous does the virus protection software need to be? Will extensive firewall protection be required? What, if any, hardware security devices should be included? What measures will be taken to protect the network from physical damage from natural disasters such as flood or fire? Will the network be hard-wired or wireless? VO: Wireless networks must be designed for optimum signal coverage and require more extensive protection against malicious hackers and viruses. Will the network be private or will it be connected to the Internet? If an Internet connection is available, will user access be restricted to certain times of the day? Is content-control software, nannyware, required to prevent access to certain Internet sites? Are there any availability/fault tolerance issues? What downtime metrics, if any, must be met? If fault tolerance is a high priority, the network will require redundant components, paths, and services to avoid single points of failure. What future planning issues should be considered? What level of scalability, if any, is required for future expansion? What form might this future expansion take? More users? More required throughput? More disk storage capacity? Are there any issues around proprietary hardware or software that must be considered? Is the business restricted for some reason in its choice of vendors? A requirement to use (or not use) a specific vendor can impact the selection or availability of certain required components and can affect component compatibility. What are the scheduling parameters for this project? When is delivery of an installed and operational network expected? If the deadline is short, you will need to consider delivery lead times when specifying hardware components. Will there be any restrictions on system availability during the installation process? (this could also be a checklist of questions put in a table or spreadsheet form) **Remember, you do not want to overwhelm the key decision maker with too many technical questions, instead ask the internal IT Manager/Specialist some of these questions and develop a relationship at a high level with the decision maker.** Note that some of these questions might not be pertinent to all network design projects. Neither is this an exhaustive list of everything you might ask your network decision makers. Much depends on the size, nature, and complexity of your customer's business. When interviewing the decision makers, be aware that they might not be able to readily answer all the questions you ask. Some issues such as future expansion or a downtime metric might not be something they have considered before. Budget, in particular, can be tricky. Some customers want to know how much something will cost before committing to a dollar amount. If this is the case, you will have to remain flexible and perhaps present several configuration options based on what the decision makers have indicated are priority features.

What to Ask End Users The primary need of all network end users, regardless of their role in the company, is consistent application availability, as determined by response time, throughput, and reliability. Therefore, you should ask questions that let you quantify and evaluate these requirements. To that end, you should ask end users the following types of questions about each separate task to be supported on the network: Network Needs What specific applications will they use for this task? How many transactions or output files will they typically create during some set time period? This information helps you identify the data flow and throughput requirements of the network, as well as how much data storage capacity will be required. What type of devices will they use for this task? In addition to their normal computing platform (desktop, laptop, mobile device, etc.), will they use any special devices such as network printers and scanners, medical devices, or process control instrumentation? How frequently will they use these devices during the course of their day? When will they perform this task? What shift do they work? When will they typically log onto the network? When will they log off? Will they ever access the network after hours? If so, how frequently? Will they perform this task constantly throughout the day or only intermittently? If intermittently, will it be random or at scheduled/predictable times? This information will help you identify the need for any type of load balancing service. How much and what type of data will they generate while performing this task? How much of that data will need to be stored on network devices? How much of that data will be time-sensitive, requiring fast throughput? Are there any security issues requiring secure or encrypted data transmission? Ask these and other pertinent questions about each task to be performed on the network. Do not forget to ask about basic support tasks that users tend to take for granted. Services such as e-mail, system backup, and task management software can be easily overlooked as design requirements because everyone just assumes they will be included. Assume nothing. Ask.

*base 2 or binary *

a number system containing only 2 digits 0 & 1

*video card*

a card of set of circuitry to adapt the CPU's visual output to a monitor display

*sound card*

a card or set of circuitry that converts analog sound into digital impulses and vice versa

*2. decision maker*

a company owner or manager who determines the purpose of the network and the ground rules for how it will be created and used

*host*

a computer or member of a network

*6. load balancing*

a computer processing technique for distributing workload across a network

*Which of the following would be a good reason for performing a written survey of the network stakeholders?*

a description of what the network end users will need, grouped according to each task performed

*switch*

a device that splits a single network connecting cable into multiple connection; switches, as opposed to hubs, manage the flow through of the data that they split

*modem*

a digital to analog (and vice versa) conversion device that translates digital communications of all types into sound

*classful addressing*

a division of IP addresses into classes defined by the "most significant bits" in the address

*Which of the following websites would most likely need to be load balanced with collocated web servers?*

a food fanchise's website

*3. downtime*

a measure of the time that the network is unavailable for use, expressed as an amount of time within an interval ("x minutes per week")

*12. stakeholder*

a network user who has a special interest, or stake, in how the network is designed

*peer-to-peer*

a network with no centralized authentications; data resource access is managed by each individual host

*volatile memory *

memory that is subject to loss of data without an electrical current

*payload*

actual bits of data used

*response time*

adequate use of time for user satisfaction includes image quality

*token ring*

an early networking topography characterized by a physical ring of hosts around which a packet token would travel

*thin BNC*

an early networking topography that used coaxial cable

*ethernet*

an error checked protocol that means packets contain a check-sum, which is sent along with the packet; if the sending device and receiving device check-sum do not match the packet will be re-sent until they do.

*checksum*

an error correction method where the number sent must equal the number received

*Some quiet password breaking schemes involve a(n) ____ using only a few passwords attempts a day.*

attacker

*syslog*

aystem log (an automatic log, usually written on a separate host where logs can be stored, archived and monitored)

*local area networks*

being within one physical location, often within one IP addressing subnet (LAN)

*network adaptor*

cards or sets of circuitry that allows network data exchange

*topologies*

configurations of hosts along a group of networked computers

*packets*

contain control information as well as user data

*8. nanny ware*

content-control software; software designed to control what content a reader can view, especially when it is used to restrict material delivered over the Internet

*CRM*

customer resource management

*datagram*

data portion of a packet

*IP addresses are the element that tells the router what to do with the of the packet.*

datagram

*Which of the following are the two groups of people that make up network users?*

decision makers and end users

*A WAN (wide area network) should be connected at the of a system*

demearcation

*network security*

determines the level of access and the method of firewall necessary for the client's needs

*dedicated circuits*

direct lines called T1s (or a half Ts), or ISDN (Integrated Digital Services Network)

*DNS*

dynamic name service; a name to IP address conversion service

*Which of the following is an example of a generic network support service?*

e-mail

*An office with a small number of users will require a higher number of servers and switches than one with a large number of users.*

false

*General data services such as file management and FTP transfers typically require high bandwidth.*

false

*Routers are digital to analog conversion devices that translate digital communications of all types into sound.*

false

*Star topology is characterized by broadcast messages sent by hosts wanting to communicate.*

false

*There are four types of major network architectural models.*

false

*FTP*

file Transfer Protocol

*clean power*

filtered consistent power

*5. fire wall*

hardware or software used to help keep a network secure by analyzing incoming and outgoing data according to a predefined set of rules and determining whether the data should be allowed through

*Which of the following is a difference between switches and hubs?*

hubs do not engage in any kind of computing while switches do

*HTTP*

hyper Text Markup Language

*wide area networks*

including many physical locations and multiple subnets within one office or your home; often thought of as being synonymous with the Internet; local area networks are often thought of as being within one office or your home; abbreviated WAN

*router*

inspects IP address, and in some cases packet content, and routes it to its destination by way of the quickest most efficient path

*client/server network*

is more stable than peer-to-peer and provides a higher level of network security and data protection

*Which of the following is a way for computers to improve cleaner power?*

making more powerful rechargeable batteries

*Proper management and ___ are the only way to keep a network running.*

monitoring

*The central processing unit (CPU) is mounted to which of the following?*

mother board

*peer-to-peer network*

network requires less hardware and are cheaper to buy and operate

*megabytes*

one million bytes (approximately - actually 1048576)

*International Organization for Standards*

organization that created the OSI model (ISO)

*control data and user infromation are contained in*

packets

*Which of the following is a problem for peer-to-peer networks compared to client/server networks?*

peer-to-peer requires at least two hosts to be available for a connection to be made

*digital phone service*

phone service that usually connects directly to the router

*GoToMyPC or LogMeIn*

programs that allow you to access your personal computer over the wide area

*An operator handled connections on an old telephone system; the exchange of data between computers is governed by a *

protocol

*Virtual Private Networks*

provide an encrypted "tunnel" through the public Internet for secure access; abbreviated VPN

*Which of the following is an example of a function of a firewall and not of a regular router?*

provide security from malicious software

*brown outs*

reduction in line voltage on the power grid usually from an overload

*device connectivity*

refers to the ability of two network entities—devices and/or processes—to make and maintain a connection

*A network that would not be affected by one failing host would use which topology?*

star

*Ring topology could make a hybrid with which other topology in order to improve its ability to isolate problems in the network?*

star

*Which of the following hardware is used to split a single network connecting cable into multiple connections?*

switch

*11. scalability*

the ability of a network to perform under an increasing or expanding workload; also the ability of a network to be successfully expanded in the future

*subnetting*

the act of using a subnet mask to limit the number of possible addresses in a TCP/IP network

*logical component*

the addressing component of networking (and all protocols and mechanisms relating to addresses)

*internet protocol*

the addressing scheme used on the Internet (IP)

*13. throughput*

the average rate at which data is transmitted across a network

*octets*

the first of four 3-digit numbers separated by points in the IPV4 addressing scheme

*Which of the following describes the User Needs subsection of the Results in the Customer Needs Report?*

to define the network requirements before making any technical design decisions

*TCP/IP*

transmission Control Protocol over Internet Protocol; the entire definition of the process of IP addressing

*A network map is used to understand how the hardware and software components of a network fit together.*

true

*6 Virtual Private Networks provide security through the public Internet using an encrypted *

tunnel

*Which of the following would be a factor when determining the required transfer rate for a device?*

user competition

*invasive carrier interferences *

viruses and other malware sent along the carrier wave frequencies that accompany electric current

*What are invasive carrier interferences?*

viruses sent along the carrier wave frequencies of electric currents

*firewall*

wall of protection sometimes provided by the ISP's cable modem or router