remote management of switches

Many data centers adopt a leaf / column architecture, which eliminates the level of aggregation. In this project, the servers and storage are connected to the leaf switches (edge ​​switches) and each leaf switch connects to two or more pins (cores) of the switch. This minimizes the number of hops that data has to get from the source to the destination and, therefore, reduces transit time or latency.

Some data centers establish a mesh or matrix network design that shows each device on a single large switch. This approach minimizes remote management of switches latency and is used for demanding applications such as high performance computing  in financial services or engineering.

Not all networks use switches. For example, a network can be organized in a token ring or connected via a bus, hub, or repeater. In these networks, each connected device sees all the traffic and reads the traffic directed to it. A network can also be established by connecting computers directly to each other, without a separate layer of network devices; This approach is mainly interesting in  contexts where latencies of less than 5 microseconds are desired and can be quite complex to design, connect and manage.

Comments

Popular posts from this blog

cloud managed switch

what does a network switch do

ms access switches