Unfortunately, our networks can’t always maintain high levels of performance, due to specific reasons. It is always important to make sure that our network could perform with stable reliability. With proper maintenance, it is possible to ensure a reasonable performance, but we should consider a number of factors that could affect the performance of our networks. Ethernet and other packet-based networks require regular maintenance. Although electrical components don’t wear our like mechanical ones, software could eventually become slower due to accumulation of digital content. An Ethernet could be connected to an ASDL plan, but without enough maintenance, the overall performance could eventually degrade. Regardless of what we do, we need to make sure that there is always enough capacity to accommodate peak demands that could happen on specific times of the day. Jitter, packet loss and latency could significantly affect VoIP and voice delivery.
Reliable VoIP connections require low jitter and low latency. There should only be moderate packet loss and a network doesn’t always have the ability to manage large number of small data packets associated with VoIP connection. It should be noted that VoIP traffic could traverse through the whole network and perhaps other networks. Traffic could be routed to various equipments resulting in a number of unexpected situations. It could really mean that packet loss, jitter and latency will be twice higher than those claimed by the service provider. On the other hand, video streaming may not require low latency, because many of these videos are stored for more than a few days and are not related to real-time requirements. It is important that our network can be extended and its connectivity could expand in many different areas. This is especially important if we want to consolidate our expanding network, especially if some sub-networks are located in remote locations. Regardless of our requirements, we should always make sure that latency and jitter will be quite low.
We should always try to get clear indications about network performance. Core areas of our network should be carefully examined and we should make sure that they could perform really well. As an example, we should always consider about Ethernet latency and it is essentially the time needed for data packets to reach the destination, especially if they need to traverse through the network. Some providers use metrics like “round trip delay” and it is the time needed for data packets to circle throughout the network and it roughly equals to double the latency figure. We often measure latency in milliseconds and although it sounds brief enough, higher latency could cause the whole network feels sluggish, especially if many people need to send large amount of data. Obviously, latency isn’t constant and jitter can be used to define how much latency could vary within a network. For example, it could tell us about the gap between the longest and the shortest latency time. Excessive latency could also indicate that the network is less stable and reliable. Packet loss also need to be minimized, so senders won’t need to re-send lost packets.