Top Three Network Considerations for Large Scale Cloud Deployments

Cloud Computing, , ,

The rapid pace of change in data and business requirements is the biggest challenge when deploying a large scale cloud.  It is no longer acceptable to spend years designing infrastructure and developing applications capable to cope with data and users at scale. Applications need to be developed in a much more agile manner, but in such a way that allows dynamic reallocation of infrastructure to meet changing requirements.

Choosing an architecture that can scale is critical. Traditional “scale-up” technologies are too expensive and can ultimately limit growth as data volumes grow. Trying to accommodate data growth without proper architectural design, results in un-needed infrastructure complexity and cost.

The most challenging task for the cloud operator in a modern cloud data center supporting thousands or even hundreds-of-thousands of hosts is scaling and automating network services.  Fortunately, server virtualization has enabled automation of routine tasks – reducing the cost and time required to deploy a new application from weeks to minutes.   Yet, reconfiguring the network for a new or migrated virtual workload can take days and cost thousands of dollars.

To solve these problems, you need to think differently about your data center strategy.  Here are three technology innovations that will help data center architects design a more efficient and cost-effective cloud:

1.  Overlay Networks

Overlay network technologies such as VXLAN and NVGRE, make the network as agile and dynamic as other parts of the cloud infrastructure. These technologies enable automated network segment provisioning for cloud workloads, resulting in a dramatic increase in cloud resource utilization.
Overlay networks provide for ultimate network flexibility and scalability and the possibility to:

  • Combine workloads within pods
  • Move workloads across L2 domains and L3 boundaries easily and seamlessly
  • Integrate advanced firewall appliances and network security platform seamlessly

2.  OpenStack

OpenStack has become the most popular open cloud platform.  A primary reason for its growing popularity is because OpenStack reduces the cost of running a data center. The area of cost savings is in licensing costs for virtualization and ongoing maintenance.  OpenStack increases operational efficiency and adheres to open standards. These open standards are key business drivers for the adoption of OpenStack in production deployments.

OpenStack also provides support for overlay networks.  In fact, to achieve true flexibility and scalability it requires overlay networks that can be efficiently deployed and managed. The “Neutron” plugin is a networking abstraction layer for OpenStack that offers such integration, allowing a number of open source tools and commercial products to act as networking back-ends.

3. Hypervisor Hardware Accelerators

There are several critical hardware offload engines that accelerate hypervisor tasks and thereby increase cloud efficiency.  As described above, overlay networks can deliver significant benefits in scalability and security of virtualized networks.  Yet, there is an unfortunate side effect. The addition of overlay networks results in significant additional packet processing, which consumes CPU cycles and degrades network performance.

Overlay networks utilize an additional encapsulation layer, and thus traditional NIC offloads cannot be utilized. The result is the consumption of large amount of expensive CPU processing resources as well as a significant reduction in network throughput.

In order for overlay networks such as VXLAN to be of real value, the resulting CPU overhead and networking degradation must be eliminated. This can be achieved by offloading the overlay network processing to the next generation of overlay-aware NICs which include hardware accelerators embedded within the network controllers.  This acceleration includes:

•   Performing checksums on both the outer and inner packets

•   Performing large segmentation offload (LSO)

•   Handling virtual machine steering (ex: NetQueue) to ensure that traffic is distributed between different CPU cores in the most efficient manner.


The Mellanox Solution

Mellanox interconnects provide a solution named CloudX.  This technology allows cloud users to enjoy the automation and flexibility of an overlay solution fully integrated into OpenStack without compromise.  This CloudX solution achieves wire-speed traffic processing rates and improved CPU utilization, even at scale.


One example of such end-to-end solution is the PLUMgrid IO Visor™ and Mellanox ConnectX®-3 Pro integration. The combined solution delivers an industry-leading traffic processing rate of 40 Gbps per server as can see in the below graph.


Eli Blog 052814


To learn more about this solution read: “Achieving a High-Performance Virtual Network Infrastructure with PLUMgrid IO Visor™ & Mellanox ConnectX®-3 Pro”.


eli karpilovski
Author: Eli Karpilovski manages the Cloud Market Development at Mellanox Technologies. In addition, Mr. Karpilovski serves as the Cloud Advisory Council Chairman. Mr. Karpilovski served as product manager for the HCA Software division at Mellanox Technologies. Mr. Karpilovski holds a Bachelor of Science in Engineering from the Holon Institute of Technology and a Master of Business Administration from The Open University of Israel. Follow him on Twitter: @Elikarpilovski.



Comments are closed.