CIS 321 Ch.3 TYU

3c) You are working at an access point with 20 other people. Three are doing a download at the same time you are. The rest are looking at their screens or sipping coffee. The access point you share has a rated speed of 150 Mbps and provides a throughput of 100 Mbps. How much speed can you expect on average for a download?

(Answer: 25 Mbps)

6. a) What are service level agreements?

a contract between a service provider and its internal or external customers that documents what services the provider will furnish. They measure the service provider

10e) What is the advantage of each compared to the others? (Priority)

All switches and routers can implement it easily. It is more efficient comparing to Overprovisioning method. It does not increase the capital expense.

Throughput

Defines amount of transfer particular amount of time. That is, it is the amount of data that actually gets transmitted from the user PC, through the Internet to the web server and back in a single unit of time. It is used to measure the performance of Internet and network connections. The throughput is usually measured in m per second (bps), and sometimes in mpackets per second. While user downloads files, the number mentioned in the progress bar, is throughput. Throughput is never constant and varies with varying time intervals and is dependent on the current network traffic and at peak times, the speed may be significantly decreased. It can also be modified by the speed of websites the user visit. Throughput will obviously always lower than the line speed.

2b) For the HDTV program in Figure 3-3, which of the speeds shown will allow real-time streaming?

FIGURE 3-3 Application Download Times at Various Transmission Speeds

3b) Distinguish between individual and aggregate throughput.

Generally, while talking about throughput, users mean about the aggregate throughput. The aggregate throughput first increases with the number of users and then becomes constant after a finite number of users. Whereas the individual throughput decreases with, increase in the number of users.

6g) What happens if a carrier does not meet its SLA guarantee?

If a carrier does not meet the guarantee of service level agreement, then carrier must faces penalties. The whole network performance is measured with all the parameter like speed, latency etc. If the weekly or monthly analysis create a very minimal percentage of time, then the penalties charged according to the agreement.

10f) Compared to e-mail and VoIP, what priority would you give to network control messages sent to switches and routers? (The answer is not in the text.) Explain your reasoning.

If there is chronic traffic load which exceeds the network capacity, user would use Priority level approach. It is because, it manages traffic based on the tolerance for delay and loss. • It can be used for reducing the impact of delays for latency intolerant traffic by sending the packets on priority basis. • Here, it gives higher priority to those frames and packets which are necessary at the destination level and others are given low priority. • The high priority frames and packets are sent immediately while the less priority frames are made to wait until the momentary congestion clears. • Also, all switches and routers can implement it easily.

10e) What is the advantage of each compared to the others? (QoS guarantees)

In Quality of Service guarantees, company should allocate reserved capacity for some traffic. The traffic which has reserved capacity gets through without any congestion. It can be used w here there is overloading due to congestion

6j) A business has an Internet access line with a maximum speed of 100 Mbps. What two things are wrong with this SLA?

In a business, if SLA mentioned the maximum speed of 100Mbps, then two wrong thing is maximum speed and related latency. SLA mainly use for the w orst case condition. No one is bother about the highest speed, rather customer bother about the minimum speed. So maximum speed is not mentioned. And the maximum speed must provide very low latency. That means totally inverse method is applied for the speed and latency. If the speed, gain latency is also affected and it is very hard to maintain a consistent latency.

10e) What is the advantage of each compared to the others? (Overprovisioning)

In overprovisioning method, there is no requirement for additional labor burden. In overprovisioning method, there is no requirement for additional labor burden. It avoids almost all momentary traffic. It can be used in LANs, as installing additional capacity is inexpensive.

Rated Speed

It is a network's rated speed. It should complete according to the merchant titles or on the standard that defines the technology. It is simply the maximum amount of data that can pass through the line per second. It is expressed in bits per second with common values are usually expressed in thousands (kilobits per second), millions (megabits per second), and in billions (gigabits per second). It depends upon number of factor like server efficiency, type of modem, network conjunction, line quality, the protocol using to send data, whether the broadband micro-filters are installed exactly. For example, Ethernet networks with rated speed 100 Mbps or 1 Gbps are referring to the maximum amount of data that can pass through the cable per second.

6f) Would an SLA guarantee specify a highest jitter or a lowest jitter?

Lowest jitter, SLA must not directly mention the jitter value, but it is closely related with other parameter like availability, speed and latency. • Jitter is measured in terms of the variability that means the average arrival time is divided by average latency. So, if latency is minimum, then jitter is also minimum. In clear ways, jitter should be minimum for better network delivery.

6c) Would an SLA specify a highest speed or a lowest speed?

Lowest?

1. a) What are QoS metrics? (Do not just spell out the acronym.)

Measures the performance of the network quantatively. Several companies' use a number of QoS metrics to quantify their quality of service. Hence, they set targets and fIx if they have met those targets. These companies are concerned with network quality of service metrics.

6i) Does residential ISP service usually offer SLA guarantees? Why or why not?

No, residential ISP services does not provide any SLA services. In very rare cases, it may be provided. • In commercial network, there were lots of module for the network engineering service. Without proper internet service, protocol must be affected. So in this case the SLA is managed with lots of costs and price. • In contrast, the residential network requirement is different and cost should be minimal. Individual customer is not agree to pay this extra burden. That why the SLA is not provided to the residential network.

Some examples of QoS metrics are as follows:

Performance Example, user-perceived latency, server throughput. Degree of fauIt-tolerance °Completeness Example, how much of relevant data is used to answer query. Accuracy Example, of a computation or decoding/encoding process. °Capachy Example, admission control limits, access to non-essential services.

Individual Throughput

The individual throughput is the share of the individuals in the aggregate throughput. It is the share that a single user will receive when many users are connecting to the same terminals in a network.

10b) How long do momentary traffic peaks last?

The momentary traffic peaks occur when there is a strong random component in the network traffic. • These momentary traffic peaks exceeds the network's transmission link capacities. • They last for a fraction of a second or a second or few seconds.

6h) If carrier speed falls below its guaranteed speed in an SLA, under what circumstances will the carrier not have to pay a penalty to the customers?

The physical layer problem to the customer end is not bonded by the carrier. That means in case the security alert or firewall management is not properly implemented in the customer end, then service may be disrupted. Sometimes, lot of special service like VLAN etc. affect the internet connection. So without any authorized service, if a customer run a special protocol, then SLA must not pay any penalties.

2. a) Is transmission speed usually measured in bits per second or bytes per second?

Transmission speed is normally measured in bits per second (bps). As it is known that bit is the basic unit of computer through which every other units of measurements can be represented. Whereas, sometimes speed is measured in bytes per second (bps) but only for file transfer.

6b) Does an SLA measure the best case or the worst case?

Worst case

10. a) Distinguish between chronic lack of capacity and momentary traffic peaks.

Usually, network capacity is higher than the network traffic. Sometimes, the network traffic posses a strong random component. When this randomness occurs, there will be occasional traffic spikes which are called as Momentary traffic peaks • These momentary traffic peaks exceeds the network's transmission link capacities. • These exist for a fraction of a second or a second or few seconds. • This cause latency because switches and routers have to store the packets and frames using buffers when they are waiting to send them out. • Buffer size is used to be small which causes loss of packets.

5d) In what units is latency measured?

in milliseconds

5c) What is latency?

refers to the numerous kinds of delay in communication processing. In networking it is defined as the time delay in the packet delivery for a particular network.

The three key points of jitter are?

successive pulses the signal amplitude phase of periodic signals.

5e) What is jitter?

the differentiation of actual periodic signal as compared to pre-defined signal in electronics and telecommunications transmission purpose.

5. a) What is availability?

the term for actual use of the network transmission to the end user people. That means the full frequency of network delivery is measured in time of unit. So, it is the percentage of time in which the network is fully operate at user end. The efficient network time delay is very less according to the full latency of the network.

Aggregate Throughput

the total throughput available to all users who shares it. It is the sum of the data rates that are delivered to all terminals in a network

5b) When should you measure error rates? Why?

• In case of network breakage the error rate should be checked for inter-networking problem. This breakages create a time delay in the delivery. • Because the flow control is major component in gaining the network throughput. The error rate is measured for the full fledge undisrupted delivery of packet to the end user.

10c) What two problems do they create?

• Latency and Packet loss - Switches and routers store the packets and frames using buffers when they are waiting to send them out. This is because of the momentary congestion and it leads to latency. Buffer size is used to be small which causes loss of packets. • Overload - There are few applications which use TCP at the transport layer. These applications used to re-transmit the lost data. But the re-transmission procedure increases the network traffic volume which leads to overloading.

5f) Why may adding applications that cannot tolerate latency and jitter be expensive?

• Latency and jitter is dependent on the particular context of the network topology and the condition of traffic. • Latency and jitter are very extremely in the networking for the unblocking network switches.

10d) What three choices do you have for reducing the impact of delays on latency-intollerant traffic?

• Priority level approach can be used for reducing the impact of delays for latency intolerant traffic. • Here, priority level is assigned to the frames and packets depending on their tolerance for latency and loss and it work as follows: 1) Voice over IP: • Voice over IP (VoIP) is extremely latency intolerant. • If there is congestion, the latency-intolerant traffic such as voice is given the higher priority and get through immediately without any delay. • The lower priority traffic waits till the higher priority traffic is sent. • Hence, delay in the transmission causes difficulties during conversation. 2) E-mail: • In contrast, E-mail is latency tolerant traffic and gets lower priority. • It can tolerate a delay of several seconds and hence, waits until the momentary congestion clears.

6d) Would an SLA specify a highest availability or a lowest availability?

• SLA all-time mention the low est availability matter. Because network should be available to all for maximum time and this thing should be measure by the percentage of calculation process. • For 100% availability, no one ask any question, but there must be a threshold for the lowest available time. If this condition braked, then service provider are ready to pay penalties.

6e) Would an SLA specify highest latency or lowest latency?

• SLA must specify the higher latency value. Because, there must be a threshold for the maximum time delay. • So in any point the latency must not cross this mentioned value. The network is more stable if the latency value is minimum at all the time.