Week 2: The Network Layer

In computer logic, what represents true and false?

1 and 0.

What are the four columns of a basic routing table?

1) Destination network. 2) Next hop. 3) Total hops. 4) Interface.

What are the basic steps of routing?

1) The router receives a data packet from one of its interfaces (it has an interface/separate IP address on each network). 2) The router looks at the destination IP. 3) The router looks up the network that ties to the destination IP, using the routing table. 4) The router forwards the traffic to the destination, or as close to it as possible.

What are the 3 primary ranges of non-routable address space?

10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16. Free for anyone to use in their internal networks.

What is the minimum length of an IP header?

20 bytes.

An 8-bit binary number can represent how many decimal numbers?

256. (2^8) Specifically, the numbers 0-255. The base 2 is why we do 2^8.

How long are the source and destination IP address fields?

32 bits.

What are IP addresses?

32-bit long numbers made up of 4 octets, which can represent 0-255.

What are subnet masks?

32-bit numbers that are normally written as four octets in decimal. They help us calculate subnet IDs.

What is the Protocol field?

8-bit field with data about what transport layer protocol is being used (usually TCP or UDP).

What is the identification field?

A 16-bit number that's used to group messages together.

The IP datagram also has a payload section. What does this contain?

A TCP or UDP packet!

What is the header checksum field?

A checksum of the contents of the entire IP datagram header.

What is an autonomous system in networking?

A collection of networks that fall under the control of ONE network operator (i.e. ISP).

What is the IETF (Internet Engineering Task Force)?

A community that works on the standards for the Internet.

What is a router?

A network device that forwards traffic depending on the destination address of that traffic.

What is ARP (Address Resolution Protocol)?

A protocol used to discover the destination hardware (MAC) address of a node with a certain IP address.

What is the padding field?

A series of zeros to make sure that the header is the correct size.

What is the opposite of a dynamic IP address?

A static IP address, usually reserved for servers and network devices.

How do subnet masks work?

A subnet mask is a binary number that has 2 sections: the beginning part, aka the mask, which is a string of 1s, and the last part which is just zeros (so it's 255.255.255.0). The part with all 1s (which is the subnet ID) tells us what we can ignore when computing a host ID. The part with 0s tells us what to keep. Take the part of the IP address that *isn't* the network ID, and put it next to the remaining pieces of the subnet mask. The ones with the 1s are subnet ID. The ones with 0s are the host ID. The size of a subnet is defined by the subnet mask. Whichever part has the 0s *in binary* shows us how many bits we have to use for a host ID. 8 bits means we have 256 possible numbers, but in practice it's 254 because 0 and 255 are usually reserved.

In computer science terms, what is a list known as?

A vector.

What is a subnet mask?

A way for a computer to use AND operators to *determine if an IP address exists on the same network.* The host ID is what's left over.

What is the address class system?

A way of defining how the global IP address space is split up.

What are the 3 classes of IP addresses?

A, B, and C.

What is the TTL (time to live) field?

An 8-bit field that indicates how many router hops a datagram can traverse before it's discarded. Prevents infinite loops.

What is a distance-vector protocol?

An older standard. Routing table is sent to all the nearby routers. It just tells us information about its immediate neighbors, and how far apart it is. That way, we can figure out the most efficient path.

What is the IP options field?

An optional field, used to set special characteristics for testing purposes.

In order to do its job, how many networks does a router need to be connected to?

At least two.

What exterior gateway protocol is in use today?

Border Gateway Protocol (BGP)

How are subnet IDs calculated?

By using a subnet mask.

How do communications happen across networks?

By using routing.

How does communication happen across the Internet?

By using routing.

What do we use now instead of classes for IP addresses?

CIDR.

What does CIDR stand for?

Classless Inter-Domain Routing

What is CIDR?

Classless inter-domain routing.

What is the destination network column in a routing table?

Contains a row for each network that the router knows about, consisting of a network ID (IP address) and the subnet mask.

How do you communicate with the IP address 9.100.100.100?

Core routers on the Internet know that this belongs to the class A network. They then route the message to the gateway router for class A using the network ID (part of the IP address). The gateway router then looks at the host ID (also part of the IP address) to get the data to the proper system. Subnets make this easier because they break these large networks down into smaller pieces.

How do core routers send datagrams normally?

Core routers only care about the network ID and use it to find the gateway router. The gateway router figures out how to send it either to a) the destination machine, or b) the next router along the path. Then the host ID is used by the final router to deliver the datagram to the proper destination machine.

What are the ranges for class D and E IP addresses?

D = 224 - 239, used for multicasting, E = 240 - 255, only used for testing purposes.

IP addresses assigned through Dynamic Host Configuration Protocol are known as what?

Dynamic IP addresses, usually reserved for clients.

What are link state protocols?

Each router advertises the LINK to each of its interfaces. So each router in the system knows every detail about the other routers in the system.

What are data packets at the Ethernet layer called?

Ethernet frames.

What is a gateway router vs. a core router?

Gateway router = entry/exit path to a network. Core router = speaks to other corer routers.

What does the network layer pertain to?

How data crosses many networks over great distances.

What is /27 notation?

How many 1s we have in the subnet mask before getting to the 0s, in binary form. This provides a nice shorthand to give us a full IP with subnet mask included.

What is the interface column in a routing table?

How the router knows which interfaces it should route traffic out of.

What is the solution for node communication in larger scopes? MAC doesn't scale well, so we can't use that.

IP addressing in the network layer.

What is a data packet at the IP level called?

IP datagram. (Comprised of a highly structured series of fields that are strictly defined.)

How do you extract your network ID using the subnet mask?

If you perform an AND between our IP address (in decimal form) and our subnet mask, you get the network and subnet ID portion of the IP address only. Now, the computer can compare this result with its own network ID.

What is the flags field?

Indicates if a datagram is allowed to be fragmented / has already been fragmented (split up into smaller datagrams)

What are the 2 categories of routing protocols?

Interior gateway protocols and exterior gateway protocols.

What is the fragmentation (fragment) offset field?

It contains values used by the receiving end to reassemble fragmented packets.

What is a demarcation point?

It describes where one network or system ends, and another begins.

What is the total hops column in a routing table?

It helps the router choose the best path.

What is CIDR in-depth?

It uses subnet masks to demarcate networks -- in other words, to describe where they begin and end. The network and subnet IDs are combined into one. This is the same thing as the slash notation, aka CIDR notation. Abandons the concept of address classes entirely. This allows for varying network sizes. And only one entry needed in the routing table (i.e. for /24 instead of for 2 Class C slots.)

MAC addresses are used for nodes communicating in which types of networks? Why?

LANs, because of the small scope.

What are the two categories of interior gateway protocols?

Link state routing protocols and distance-vector protocols.

What is NAT?

Network Address Translation: Allows computers on non-routable address space to communicate with other devices on the internet.

When do ARP table entries expire?

Often after a short amount of time, to ensure that changes in the network are accounted for.

What is the most common link state protocol?

Open Shortest Path First (OSPF)

What is non-routable address space?

Ranges of IPs set aside for use by anyone, which cannot be routed to -- No gateway router will attempt to forward traffic to them. So they work inside of individual networks, but not outside them.

What is RFC?

Request for Comments. Helps people agree on standards.

What are the most common distance bettor protocols?

Routing Information Protocol (RIP), and Enhanced Interior Gateway Routing Protocol (EIGRP).

What are routing protocols?

Special protocols that routers use to speak to each other, and share information with each other. This is how information travels across the world.

What is the next hop network column in a routing table?

The IP address of the next route that should receive data on the way to the destination.

How do these IP datagrams fit into Ethernet frames?

The data payload section of an Ethernet frame *IS* the IP datagram! This process is known as encapsulated.

What are class B IP addresses?

The first 2 octets are used for the network ID, and the second 2 are used for the host ID.

What are class C IP addresses?

The first 3 octets are used for the network ID, and the last one is used for the host ID.

How do you know which class an IP address is?

The first digit. A = 0-126, B = 128 - 191, C = 192 - 224.

What are class A IP addresses?

The first octet is used for the network ID, and the last 3 are used for the host ID.

What is the second field in an IP datagram?

The header length. Almost always 20 bytes in length with IPv4. But the field itself is 4 bits.

What are the 2 primary sections of the IP datagram?

The header, and the payload.

What is the max size of a single datagram?

The largest number you can represent with 16 bits: 65,535. If the data needed to be sent is larger, then it needs to be split into individual packets.

What are the 2 sections of an IP address?

The network ID and the host ID.

What happens when there isn't an entry for the requested IP address in the ARP table?

The node that wants to send data *sends a broadcast ARP message* to the MAC broadcast address, which is all FFs. All computers receive this message. When the computer with the proper IP address gets this message, it sends back an ARP response. This response contains the MAC address. Now, we know the destination MAC address. It also puts the address into the ARP table.

What is the base 10 numeral system?

The numeral system that uses the numbers 0 - 9 as its base.

What is subnetting?

The process of taking a LARGE network and splitting it up into many smaller networks (subnets).

What is the third field in an IP datagram?

The service type field. 8 bits. Specifies details about quality of service (QoS) technologies.

What is the 4th field in an IP datagram?

The total length field. For the total length of the IP datagram.

What is the first field in an IP datagram?

The version field. 4 bits. IPv4 or IPv6?

What are interior gateway protocols?

They are protocols used by routers to share information within a *single* autonomous system.

What do address classes (such as A, B, and C) do?

They break up the global IP space into discrete (sub)networks.

What is the IANA (Internet Assigned Numbers Authority)?

They manage IP address allocation, and ASN allocation.

Where in the IP address can you find a subnet?

They take some bits from the host ID as part of the IP address.

What is the number 1 goal of core internet routers, and why?

To get data to the edge router of an autonomous system, because that gets us to a well-defined network that we can work with.

What are exterior gateway protocols?

Used for the exchange of information between independent, autonomous systems. They share information across different (external) organizations.

What is an ARP table?

Used in most networking devices. It's a list of IP addresses and the MAC addresses associated with them.

What is the binary (base 2) system?

Uses only 0 and 1, because of the on/off state in computers.

How are IP addresses distributed?

Usually to organizations and companies, instead of being determined by hardware vendors. For example, all of IBM's addresses start with 9, and IBM owns every single address that starts with 9.

What are operators?

What computers use to make calculations, such as addition, OR, and AND.

What is ASN (Autonomous System Number) allocation?

assigning numbers to individual autonomous systems. 32-bit numbers, but not split up by dots. IBM has its own ASN, as do other companies.

IP addresses belong to...

networks, not the devices attached to them.