Virtualization and Cloud Computing is Changing the Network to East-West Routing

It’s  called “east-west” networking, which when compared to its predecessor,  “north-south” networking, evinces images of maelstroms and hurricane  winds and tsunamis for some reason. It could be the subtle correlation  between the transformative shift this change in networking patterns has  on the data center with that of El Niño’s transformative power upon the  weather patterns across the globe.

Traditionally,  data center networks have focused on North-South network traffic. The  assumption is that clients on the edge would mainly communicate with  servers at the core, rather than across the network to other clients.

But  server virtualization changes all this, with servers, virtual  appliances and even virtual desktops scattered across the same physical  infrastructure. These environments are also highly dynamic, with  workloads moving to different physical locations on the network as  virtual servers are migrated (in the case of data center networks) and  clients move about the building (in the case of wireless LANs).

-- Distributed Core And East-West Routing--The Network Is Changing (Stephen Foskett, Network Computing Magazine)

Though  the term “east-west networking” really focuses on the maelstrom of  traffic inside the data center, it is also applicable to the traffic  patterns occurring outside the data center as well. The advent of cloud  and virtualization has made it such that integration of cloud-hosted  resources is becoming more and more appealing, if not already a done  deal. That means there’s a lot more east-west networking between “data  centers” – between the corporeal data center of the enterprise and the  ethereal data center out there, in the cloud. To maintain the  efficiencies and cost-savings gained by leveraging those cloud  resources, such communications must occur, after all, lest organizations  find themselves managing two completely separate processes and sets of  policies governing the delivery of the applications they have deployed.  Even dismissing the impact of monitoring and management-related traffic,  it’s important to note that modern distributed architectures may result  in the same sort of traffic patterns for live, user-oriented traffic.

Closely related to the concept of directional networking is that of trombone networking,  a phenomenon we’ve been seeing more and more of as virtualization takes  hold of data centers the world over. It, too, is equally applicable to  multi-data center deployments:

When L2 domains stretch across multiple data centers, traffic flows belonging to a single user session might have to traverse Data Center Interconnect (DCI) link multiple times.

-- TRAFFIC TROMBONE (WHAT IT IS AND HOW YOU GET THEM)

INFRASTRUCTURE is IRRELEVANT – NOT

A  pet peeve of mine is the notion that cloud makes infrastructure  irrelevant in some way. Whether organizations themselves must deal with  the infrastructure does not make it irrelevant. It may make it less  frustrating, less costly, less troublesome on a day-to-day basis, but it  is not in any way irrelevant. The infrastructure must still exist and,  as we’re seeing in the real world, it’s not going away despite the eager  predictions of pundits that it would. The enterprise needs to retain  some measure of governance over applications, and they recognize that  this means control over the infrastructure so vital to delivering them.

But  it is important to recognize that infrastructure is changing, whether  inside or outside the data center. The architectures and networking  techniques of yesterday are not necessarily well-suited to the  architectures and applications of this afternoon and tomorrow. There are  unintended consequences to not paying enough attention to the network  infrastructure and the changes being wrought on traffic patterns by the  disruptive force that is cloud and virtualization. This is no summer  rain shower we’re looking at, we’re facing a full-on technological  squall that will, if ignored, hit the data center head on.

IT  professionals need to take the wheel and be very aware of the impact of  the change in traffic patterns wrought by cloud and virtualization – and  not just inside the data center, but outside, as well. Architectural  solutions to one challenge may inadvertently trigger a change in traffic  in the network that causes congestion, overload, or simply poor  performance. The notion of extending a VLAN across data centers, into  the cloud, sounds appealing from the perspective of managing components  in a heterogeneous deployment model, but may in fact be a source of  performance and ultimately availability issues due to trombone traffic  or east-west networking. Depending on the type of traffic bouncing back  and forth, the result could be anything from a TCP retransmission storm  to a bouncing up-down status of remote nodes in the data center caused  by a heart-beat check on the Load balancer that is too short a time when checking across a LAN to a remote site.

A  thorough understanding of networking and the infrastructure that is its  foundation as well its relationship to applications is necessary when  architecting a data center network capable of not just supporting but  adapting to future challenges that arise as a result of virtualization  and cloud computing .  A silo-based IT organization cannot effectively address the impact from  virtualization and cloud computing because no one team has all the  pieces necessary. A more collaborative approach or an approach in which a  team of cross-functional experts is at the fore will be required to  navigate the coming storms.