Those familiar with deploying virtual machines (VMs) know that in order to ensure performance, VMs must be tied to physical platforms. As the demand for data-intensive virtualized and cloud solutions continues to increase, more powerful server platforms will be required to deliver this performance without significantly multiplying hardware infrastructure for every VM.

The new Intel Sandy Bridge series (Intel's Xeon E5-2600 processor family) is ideally suited for enabling more powerful and efficient virtualized solutions for high-throughput, processing-intensive communications applications. This latest dual-processor architecture features an increased core count, I/O, and memory performance to allow more virtual machines to run on a single physical platform. Virtualization can be extremely memory-intensive, as more VMs typically require more total system memory. In order to optimize performance and easily manage VMs, each one usually requires at least one physical processor core. Using the new Sandy Bridge E5-2600 series architecture can enable individual physical servers to support greater numbers of virtualized appliances, thereby consolidating hardware for lower operational costs, preventing against VM sprawl, and simplifying transition to the cloud with opportunities for scaling up over multiple cores.

The Benefits of Virtualization
Modern carrier-grade platforms comprise unprecedented amounts of processing, memory, and network I/O resources. For developers, though, these goodies also come with the mandate to make the most effective use of modern platforms through scaling and other techniques. Through the intelligent use of carrier-class virtualization, developers can create highly scalable platforms and often eliminate unnecessary over-provisioning of resources for peak usage.

Current advances in multicore processors, cryptography accelerators, and high-throughput Ethernet silicon make it possible to consolidate what previously required multiple specialized server platforms into a single private cloud. 4G wireless deployments, HD-quality video to all devices, the continuing transition to VoIP technologies, increased security concerns, and power efficiency requirements are all driving the need for more flexible solutions.

By deploying a private cloud with virtual machine infrastructure, one's hardware becomes a pool of resources available to be provisioned as needed. The control plane, data plane, and networking can all share the same pool of common hardware.

Deployments can be easily upgraded by simply adding physical resources to the managed pool. Additionally, migrating VM instances from one compute node to another, as Figure 1 shows, can be non-disruptive.

Many telecom solutions require multiple different hardware solutions simply because they are made up of applications that run on different operating systems. In a private cloud deployment, multiple operating systems can be run on the same physical hardware, eliminating this requirement.

A private cloud enables running instances (VMs assigned to a specific function) to be tailored to different workload environments. For example, a dedicated service level can be assigned to each instance, and as demand increases or decreases, other instances can be spawned or decommissioned as necessary. This allows each process workload to be tailored for the moment-in-time demand required (see Figure 2). This ability to tailor each process workload to address moment-in-time demand means the practice of over-provisioning all resources for a "peak workload" can go by the wayside. As resources are no longer needed, they are simply added back into the pool to be used by other instances that may need to be spawned.

Virtual machines allow for the more efficient use of hardware resources by allowing multiple instances to share the same physical hardware, maximizing the use of those resources and increasing the work per watt of power consumed when compared to traditional infrastructure.

VMs also allow for 1+1 and N+1 redundancy through the use of multiple virtual instances running fewer independent hardware nodes, such as AdvancedTCA SBCs. In addition, VMs often require fewer physical nodes to achieve the same level of redundancy. By reducing the physical node count to achieve the same uptime goals, less power is consumed overall (see Figure 3).

AdvancedTCA and Virtualization
For private clouds running VM infrastructure, choosing AdvancedTCA chassis with SBCs for the compute node (the most common core element in any private cloud) makes sense because of their commonality, variety, manageability, and ease of deployment.

Network switches with Layer 3 functionality are the glue that holds the private cloud together. The selection of AdvancedTCA switches will depend largely on the internal and external bandwidth required for each compute node. Video streaming or deep packet inspection typically requires much more bandwidth (and thus higher bandwidth switches) than SMSC messaging, for example, to optimize performance.

The last necessity is also one of the most critical: shared storage. For an instance to be launched or migrated to any physical node, all nodes must also have access to the same storage. In private cloud infrastructure, a high-performance SAN and a cluster file system often supply this access. Connectivity options typically include Fibre Channel, SAS, and iSCSI connectivity. iSCSI with link speeds of up to 10 Gbps is the least intrusive approach to implementation to each node, as the SAN can be connected to AdvancedTCA fabric switches to provide storage connectivity to each node.

To avoid excessive use of fabric bandwidth for storage connectivity in high-throughput environments, employing SAS or Fibre Channels that are directly attached and connected externally to each node via RTMs is a viable option. With multiple manufacturers now making AdvancedTCA blade-based SANs as well as NEBS certified external SANs, many options are available to meet the SAN requirements for a carrier-grade private cloud.

How Sandy Bridge Processors Optimize AdvancedTCA Platforms
The new Intel Xeon processor E5 family, based on the Sandy Bridge microarchitecture, changes how well software applications run on AdvancedTCA platforms. It supports innovative networking through 40-gigabit Ethernet, and its features allow for advanced virtualization and cloud computing techniques.

The Intel Xeon E5-2600 series CPUs consist of up to eight cores, each running up to 55 percent faster than its Xeon 5600 predecessor. It can therefore deliver much higher server performance to the enterprise market. Furthermore, new enterprise servers can support up to 32 GB dual in-line memory modules (DIMMs) so memory capacity can increase from 288 GB to 768 GB using 24 slots. E5-based AdvancedTCA compute blades with more limited board real estate are expected to support up to 256 GB in 16 VLP RDIMM slots at launch. This represents a 40 percentincrease over prior blades.

Greater power efficiency is another key benefit. The E5 family provides up to a 70 percent performance gain per watt over previous generation CPUs. Communications OEMs can develop power-efficient dual processor blades for service providers that fully meet or beat AdvancedTCA power specifications.

But the real game-changer lies in the E5-2600's integrated I/O, which allows designers to reduce latency significantly and increase bandwidth. AdvancedTCA's 40G fabric has been backplane-ready since 2010 in anticipation of an updated PICMG specification release. Since then, solution providers have sought ways to eliminate bottlenecks and utilize as much of the fabric as possible. Now that Intel has integrated the new PCI-Express 3.0 with 40 lanes aboard each Xeon® processor and Quickpath Interconnects (QPIs) linking each CPU together, I/O bottlenecks are reduced, throughput is increased, and I/O latency is cut by up to 30 percent. A standard dual Xeon® E5-2600 CPU configuration offers up to 80 lanes of PCIe Gen3, which provides 200 percent more throughput than the previous generation architectures.

The overall result is much higher I/O throughput. New AdvancedTCA blades will now be able to deliver more than 10 Gb/s per node. This is a critical milestone for wireless video applications that service providers are so hungry to launch. Greater overall performance and higher performance per watt are significant by themselves, but having enough I/O capacity to match the processor capabilities makes for even greater advances in application throughput.