Full mesh, interconnected mesh, centralized and virtual switch are additional examples of emerging architectures included in the published ANSI/TIA-942-B standard. Like the Spine-and-Leaf architecture, these new options provide lower latency and higher bandwidth, and they include nonblocking ports between any two points in a data center.
- A full-mesh architecture dictates that each switch be connected to every other switch. This architecture does not scale very well Because switches are not typically in an equipment distribution area (EDA) and a fabric is not used for top-of-rack (ToR) topology. A full-mesh architecture is often used in small data centers and metropolitan networks.
[image]
- An interconnected-mesh architecture, similar to a full-mesh architecture, is highly scalable, making it less costly and easy to build as a company grows. Interconnected-mesh architecture typically maintains one to three interconnection switches (may be HDAs or EDAs), and is nonblocking in each full-mesh POD.
[image]
- In a centralized architecture, the server is connected to all fabric switches, using simple and low-latency links. Often managed by a single server, a centralized architecture is easy for a small staff to maintain. Port limitations, however, can prohibit this type of architecture from scaling well. As such, like full-mesh architecture, centralized architecture is typically used in small data centers.
[image]
- Virtual-switch architecture, though similar to centralized architecture, uses interconnected switches to form a single virtual switch. Each server is connected to multiple switches for redundancy, leading to a potentially higher latency. Unfortunately, virtual-switch architecture does not scale well unless a fat-tree or full-mesh architecture is implemented between virtual switches.
[image]
