A data center fabric is a system of switches and servers and the interconnections between them, represented as a fabric. A data center fabric allows for a flattened architecture in which any server node can connect to any other server node, and any switch node can connect to any server node.
The flattened architecture of fabrics is key to their agility. Data center fabric architectures typically use one or two tiers of switches and traffic, as opposed to data centers that implement multi-tier architectures. Traffic can be transmitted between server nodes by traversing switches, which results in extreme efficiency and low latency.
The fabric is represented by a spine and leaf design in which the fabric mesh incorporates devices on the edge (the leaves) and switches on the spine. In this architecture all nodes are only one hop away from each other, in many cases traversing only a single switch.
Cloud computing, virtualization, big data, and dynamic applications have exerted pressure on the multi-tier data center architecture, which cannot efficiently manage the volume and scale of data that new technologies process, generate, analyze, transmit, and store. Multi-tier architectures were optimized for north-south traffic—traffic from the data center to the end user—not east-west traffic. East-west traffic, or horizontal traffic, is traffic that occurs within the data center. East-west traffic dominates in nearly all cases of cloud computing, virtualization, and big data.
The data center fabric architecture model unites holistically all data center resources from processor cores to memory—servers, storage, the network, and peripherals. Its architecture is designed to handle the increase in east-west traffic while maintaining north-south traffic connectivity to end users.