May/June 2019 I 41
3-LEVEL NETWORK STRUCTURE VS. 2-LEVEL
SPINE-AND-LEAF NETWORK STRUCTURE
In contrast to the traditional enterprise, where data
center traffic is dominated by local client-to-server
interactions (north to south), the network traffic of the
large internet data center is dominated by the server-toserver
traffic (east to west) required for cloud computing
applications. The number of users accessing data via
applications is not only huge, they also have diversified
and fragmented demands that require an uninterrupted
user experience. Internet data centers require higher
bandwidth and a much more efficient network
architecture to support spikes in heavy traffic. These
spikes could be driven by anything, including video
calling, demand for online music and videos, gaming,
shopping, and news events.
The current mainstream 3-level tree network
architecture is based on the traditional north-to-south
transmission model. When a server needs to communicate
with another server from a different network segment,
its server must pass through the path of access layer ->
aggregation layer -> core layer <- aggregation layer <-
access layer. In a big data service with thousands of
Traditional
Architecture
FIGURE 1: Traditional 3-level vs. spine-and-leaf 2-level network architecture.
Spine-and-Leaf Mesh
Architecture
“Pod”
West East
South North
The market for blockchain is
set to jump from $1.2 billion
in 2018 to $23 billion by 2023.
servers communicating in a cloud computing
environment, this model is not effective as it consumes
a large amount of system bandwidth and creates
latency concerns.
To address these challenges, the world’s large internet
data centers are increasingly adopting a spine-and-leaf
network architecture, which is more effective for
transferring data between servers (east to west) as
depicted in Figure 1.
This network architecture consists primarily of two
parts: a spine switching layer and leaf switching layer.
The most beneficial feature is that each leaf switch
is connected to each spine switch within a pod, which
greatly improves communication efficiency and reduces
the delay between servers. In addition, a spine-and-leaf
2-level network architecture avoids expensive core-layer