Fibre Channel storage has imparted a traditional network philosophy and implementation methodology to IP-based storage. The dedicated
storage network has a proven design that provides performance,
predictability, and manageability to storage deployments. The storage
traffic is isolated from the application traffic to allow each network
to be optimized for its own purpose.
As a direct result of increasing east-west traffic within the data center (server-server, server-storage, etc.), an alternative to the
traditional access aggregation-core network model is becoming more
widely used. This architecture is known as a Clos or leaf-spine network
and is designed to minimize the number of hops between hosts, thus
reducing network communication latency.
In a leaf-spine architecture, shown below, the access layer is referred to as the leaf layer. Servers and storage devices connect to
leaf switches at this layer.
At the next level, the aggregation and core layers are condensed into a single spine layer. Every leaf switch connects to every spine
switch to ensure that all leaf switches are no more than one hop away
from one another. This minimizes latency and the likelihood of
bottlenecks in the network. A leaf-spine architecture is highly
scalable. As administrators add racks to the data center, a pair of leaf
switches are added to each new rack. Spine switches may be added as
bandwidth requirements increase.