When designing and ordering the compute/server layer of the VSPEX solution
described below, several factors may impact the final purchase. From a virtualization perspective, if a system workload is well understood, features such as Memory Ballooning and Transparent Page Sharing can reduce the aggregate memory requirement.
If the virtual machine pool does not have a high level of peak or concurrent usage, reduce the number of vCPUs. Conversely, if the applications being deployed are highly computational in nature, increase the number of CPUs and memory purchased. Current VSPEX sizing guidelines specify a virtual CPU core to physical CPU core ratio of 4:1. This ratio was based upon an average sampling of CPU technologies available at the time of testing. As CPU technologies advance, OEM server vendors that are VSPEX partners may suggest differing (normally higher) ratios. Please follow the updated guidance supplied by your OEM server vendor.
Testing on the release of Intel’s Ivy Bridge series processors shows significant increases in VM density from the server resource perspective. If your server deployment comprises Ivy Bridge processors, we recommend increasing the
vCPU/pCPU ratio from 4:1 to 8:1. This essentially halves the number of server cores required to host the RVMs.
Figure 19 demonstrates results from tested configurations: Overview
68 EMC VSPEX Private Cloud: VMware vSphere 5.5 for up to 1,000 Virtual Machines Enabled by Mircosoft Windows Server 2012 R2, EMc VNX Series, and EMc Powered Backup-Proven Infrastructure Guide
69 EMC VSPEX Private Cloud: VMware vSphere 5.5 for up to 1,000 Virtual Machines Enabled by Mircosoft Windows Server 2012 R2, EMc VNX Series, and EMc Powered Backup-Proven Infrastructure Guide Table 5 lists the hardware resources used for the compute layer.
Table 5. Hardware resources for the compute layer
Component Configuration
VMware vSphere servers
CPU 1 vCPU per virtual machine 4 vCPUs per physical core For 200 virtual machines:
200 vCPUs
Minimum of 50 physical CPUs For 300 virtual machines:
300 vCPUs
Minimum of 75 physical CPUs For 600 virtual machines:
600 vCPUs
Minimum of 150 physical CPUs For 1,000 virtual machines:
1,000 vCPUs
Minimum of 250 physical CPUs Memory 2 GB RAM per virtual machine
2 GB RAM reservation per VMware vSphere host For 200 virtual machines:
Minimum of 400 GB RAM
Add 2 GB for each physical server For 300 virtual machines:
Minimum of 600 GB RAM
Add 2 GB for each physical server For 600 virtual machines:
Minimum of 1200 GB RAM Add 2 GB for each physical server For 1,000 virtual machines:
Minimum of 2000 GB RAM Add 2GB for each physical server Network Block 2 x 10 GbE NICs per server
2 HBA per server
File 4 x 10 GbE NICs per server
Note: Add at least one additional server to the infrastructure beyond the minimum requirements to implement VMware vSphere High Availability (HA) functionality and to meet the listed minimums.
70 EMC VSPEX Private Cloud: VMware vSphere 5.5 for up to 1,000 Virtual Machines Enabled by Mircosoft Windows Server 2012 R2, EMc VNX Series, and EMc Powered Backup-Proven Infrastructure Guide
Note: The solution recommends using a 10 GbE network or an equivalent 1GbE network infrastructure as long as the underlying requirements around bandwidth and redundancy are fulfilled.
VMware vSphere 5.5 has a number of advanced features that help maximize
performance and overall resource utilization. The most important of these are in the area of memory management. This section describes some of these features, and the items to consider when using these features in the environment.
In general, virtual machines on a single hypervisor consume memory as a pool of resources, as shown in Figure 20.
Figure 20. Hypervisor memory consumption
Understanding the technologies in this section enhances this basic concept. Memory compression
Memory over-commitment occurs when more memory is allocated to virtual machines than is physically present in a VMware vSphere host. Using sophisticated techniques, such as ballooning and transparent page sharing, vSphere 5.5 can handle memory over-commitment without any performance degradation. However, if memory usage exceeds server capacity, vSphere might resort to swapping out portions of the memory of a virtual machine.
VMware vSphere memory
virtualization for VSPEX
71 EMC VSPEX Private Cloud: VMware vSphere 5.5 for up to 1,000 Virtual Machines Enabled by Mircosoft Windows Server 2012 R2, EMc VNX Series, and EMc Powered Backup-Proven Infrastructure Guide Non-Uniform Memory Access (NUMA)
vSphere 5.5 uses a NUMA load-balancer to assign a home node to a virtual machine. Because the home node allocates virtual machine memory, memory access is local and provides the best performance possible. Applications that do not directly support NUMA also benefit from this feature.
Transparent page sharing
Virtual machines running similar operating systems and applications typically have similar sets of memory content. Page sharing enables the hypervisor to reclaim any redundant copies of memory pages and keep only one copy, which frees up the total host memory consumption. If most of your application virtual machines run the same operating system and application binaries, total memory usage can be reduced to increase consolidation ratios.
Memory ballooning
By using a balloon driver loaded in the guest operating system, the hypervisor can reclaim host physical memory if memory resources are under contention, with little or no impact to the performance of the application.
This section provides guidelines for allocating memory to virtual machines. The guidelines outlined here take into account vSphere memory overhead and the virtual machine memory settings.
vSphere memory overhead
Some associated overhead is required for the virtualization of memory resources. The memory space overhead has two components:
The fixed system overhead for the VMkernel Additional overhead for each virtual machine
Memory overhead depends on the number of virtual CPUs and configured memory for the guest operating system.
Allocating memory to virtual machines
Many factors determine the proper sizing for virtual machine memory in VSPEX architectures. With the number of application services and use cases available, determining a suitable configuration for an environment requires creating a baseline configuration, testing, and making adjustments for optimal results.