Data Communication and Enterprise Networking Analysis Essay (Essay Sample)

Data Communications and Enterprise Networking
Student’s Name
Institutional Affiliation
Course Name and Number
Instructor
Due Date
Data Communications and Enterprise Networking
1 Introduction
When looking at computer networks, it is enumerated that data is sent in packets (small blocks). Each packet will be transmitted as a specified individual and may all together follow a different route to reach its specific destination. Upon arriving at their destined destination in conjunction with the original message, they have to be assembled to form the original message (Kompella & Bahadur, 2019). However, if the webserver is unresponsive, there is a network congestion or a technical glitch in the particular path may ultimately prevent the message from reaching the specified destination. We use the most common platforms for diagnosis of such network failures and congestions, namely the Ping and Traceroute.
Ping is a networking utility that helps one check if a given Ip address can be accessed or not. As enumerated by Kompella and Bahadur (2019), for ping to work, we send a packet to the specified address and wait for a reply. It is also used to report errors and measure round trip times. In other cases, ping is also used to check if the machines (computers) in the local network are active.
On the other hand, a traceroute is a utility that traces a packet from a designated computer to the given host, and at the same time, show the number of steps (hops) required to reach there, in conjunction with the time for each step (Kompella & Bahadur, 2019). It is also enumerated that traceroutes work by sending the packets of data with a rather low survival time (Time to Live – TTL), which gives us the exact number of steps (hops) the packet can survive before it is returned.
If a packet fails to reach the final destination and expires at the intermediate step, it is returned as a self-identified packet. So, by gradually increasing the TTL, traceroute can identify the intermediate hosts. If any of the steps (hops) have a return value of "Request timed out," it will denote network congestion and, as such, be the primary reason behind the dropping of connection and slow loading of Web Pages.
2 MTU Size
Fragmentation will only occur in the network layer when the maximum size of the datagram is greater than the maximum size of data being held, i.e., Maximum Transmission Unit (MTU). The network layer will divide the datagram from the transport layer into fragments so that it is not disrupted. Since we have 16 bits for total length, the IP datagram's maximum size is 216 – 1 = 65,535 bytes. In most cases, fragmentation will occur at the destination site's network layer and will occur at the routers. The source side does not require fragmentation due to exceptional segmentation by the transport layer; this is because instead of doing segmentation at the transport layer, and in turn handling the fragmentation at the network layer, the transport layer will query the datagram’s data limit and frame it while doing segmentation in such a way that it can fit the frame without the need of fragmentation (Cohen, 2017).
The particular receivers will identify the particular frames with identification based on the field IP header. Moreover, it is further denoted that each fragment will have a frame of the same identification number. In this case, the receiver will have to identify the fragments offsets fields in the IP header. Also, overhead at the network layer will be essential basing on the extra header introduced to fragmentation (Ashwood-Smith, 2017).
Since there is fragmentation, we also have the reassembly of fragments. This only takes place at the destination and not the routers since packets will always take independent paths (datagram packet switching), so all of the particular fragments may not meet at the router, and hence fragmentation will have to occur again (Ashwood-Smith, 2017). Another disadvantage is that the fragments may ultimately arrive in a disorganized format, which may require patching.
Based on the current networking trends and my observation, I conclude that ~1449 bytes are the largest amount of data transmitted without fragmentation. This is because I have attempted to send packets of various lengths through my hardware and, as such, run into some technical issues when the particular size reaches ~1450 bytes. However, everything seems to be fine when the packet size is less than ~1450 bytes. Following this challenge, I have debugged the NIC driver and outputted what I get right before data is sent out. We see the TCP handshake when the packet is < ~1450 bytes are sent. Once we send data over ~1449 bytes, we never see the TCP handshake or the NIC driver packet.
It is further denoted that TCP packets that exceed the particular network’s MTU will definitely get lost. According to Ashwood-Smith (2017), a critical point to note is that fragmentation will severely degrade TCP performance; to that end, when we allow packets large enough for fragmentation is rather counter-productive. Also, endpoints will ultimately handle the path MTU discovery, thereby reducing the datagram size and achieving good performance. These will be the things working for the intended notation and not cause a problem.
The endpoint will have to source for smaller packets, try them, and discover the maximum packet size that can be sent without fragmentation and settle on that. It is further enumerated that a misconfigured firewall will ultimately interfere with the MTU discovery (Tsirkin, 2018). Also, ICMP will act as an internet host requirement and blocking it indiscriminately, thereby breaking TCP.
3 RTT Versus Distance
We can define round-trip time (RTT) as the duration, usually measured in milliseconds, from when a particular browser sends a request to the point it receives a response from the particular server. It is considered a key performance entity based on web applications' metrics and one of the most critical, along with Time to First Byte (TTBF), if measuring network latency, time, and page load (Attar et al., 2018).
We typically use ping, a command-line tool that bounces a server's request and calculates the time it takes to reach a user device, just as enumerated in figure 1.0. Also, a key point to note is that the RTT may be higher than measured by the given ping command due to the network congestion and server throttling.
Figure 1.0
However, the actual RTT can be influenced by:
* Distance – This is the length the particular signal has to travel, correlating with the time taken for a request to reach a server and the response to be taken back to the browser.
* The Transmission Medium – The medium here is enumerated as the route through which the particular signal goes through (e.g., fiber optic cables, wire, and copper). This greatly impacts the speed through which the signal moves from one point to another. As such, the signal from the server is routed back to a user.
* The number of Network Hops – The intermediate servers or routers will take time to process a given signal, thereby increasing the resultant RTT. In this case, it is enumerated that the more hops a given signal will go through will ultimately result in a higher RTT.
* Levels of Traffic – We will automatically have the RTT increase when the network is congested with high traffic. Consequently, in times of low traffic, the RTT will significantly reduce.
* Response time of the Server – The processing capacity of a given targeted server will ultimately be based on the number of requests handled at that particular point in time, emphasizing how much of the server-side is at work. A longer server time response will result in increased RTT.
It is rather evident that there is a positive correlation between RTT and distance concerning the different components contributing to RTT (propagation, transmission, and queuing delays). This indicates that with an increase in the distance, the RTT also increases. This shows that the distance from which the particular ping is executed will ultimately affect the RTT (Attar et al., 2018). If we want to ensure that we have a minimal RTT in conjunction with the distance, the distance has to be of a rather smaller value; consequently, a higher value in the distance category will result in an automatic increase in the RTT.
With the current technologies, we can reduce RTT using CDNs. We can describe CDNs as a network of website content strategically placed in the form of servers (Garcia & KIRSCHBERG, 2020). In conjunction with the CDNs, we can ultimately affect the RTT using the Points of Presence (PoPs), which reduces the distance through which the signal has to travel and make hops fetch data from a given server. Web caching also plays a significant role when dealing with CDNs. Load distribution is another route for reducing RTT. Also, scalability will eliminate the use of bottlenecks, thereby reducing RTT by a significant level. Also, Tier 1 access will greatly reduce the signal's RTT.
4 Traceroute
It is denoted that traceroutes comes in handy for the tasks about network inspection. We can show the IP address of routers on the particular machine that has given the route to the destination server, similar to figure 2.0.
Figure 2.0
Domains are enumerated on the last 2 hops, in which the IP’s reverse DNS record. The reverse is also denoted in the PTR records in the format of 4.3.2.1, as illustrated. The use of many containers is not considered elegant enough, and its management...