Avoid unnecessary risk and overhead by choosing a robust and production-proven hypervisor as the foundation for your virtualized datacenter. Selecting the right hypervisor is the first step towards success in building a virtual infrastructure.
Not all hypervisors are equal. Learn more about how VMware vSphere Hypervisor is - and will continue to be - the industry's most robust and production-proven hypervisor and why VMware is the best choice for building a virtual infrastructure.
- Comparing Hypervisors
- Hyper-V and Xen Architectures: Too Much Code
- Achieve Scalable Performance
- Why File Systems Matter
- An Ecosystem of Virtualization Security Solutions
- Industry Recognition
Questions? 1-877-486-9273
Related Products
Achieve Better Scalability and Performance in your Data Center
The hypervisor plays a key part in delivering scalable virtualization performance. See detailed performance demonstrations and comparisons in the performance section of the VMware website.
Watch a technical video on the VMware Performance Advancements in VMware vSphere.
You’ll see that VMware vSphere Hypervisor achieves high-performance throughput in a heavily virtualized environment, even as the number of total supported users and virtual machines per physical host increases. Join the discussion on the latest performance topics on VROOM!, VMware’s performance team blog. Blog discussions include:
- Understanding Memory Resource Management in VMware ESX Server
- Performance of Exchange Server 2007 in a Fault Tolerant Virtual Machine
- Performance Evaluation of VMXNET3 Virtual Network Device
How fast can ESXi go? 364,000 IOPS and more!
I/O is one of the most critical performance bottlenecks in virtual environments, but even the most I/O-intense application runs fast on VMware ESXi. The result is that end users have no idea that their applications are being served from a virtual environment—any latency or overhead is usually imperceptible to the end-user. A recent test conducted by VMware with EMC showed a single ESXi host is capable of driving over 364,000 IOPS, requiring the SSD equivalent of thousands of rotating disks. To put this result into perspective, you would need to run 700,000 Microsoft Exchange mailboxes (LoadGen heavy user profile) on a single server to generate this I/O. In the same test, a single ESXi virtual machine achieved over 120,000 IOPS. With that kind of performance power available, even your most demanding workloads can be virtualized.
Why does ESXi scale and perform so much better than other vendors’ offerings? There are a number of reasons; some are articulated in a recent VMware article, “ A Look at Some VMware Infrastructure Architectural Advantages.” Two main reasons are 1) the VMware ESXi direct driver model and 2) its more effective management of memory.
Advantages of the ESXi Direct Driver Architecture for Performance
The VMware vSphere Hypervisor direct driver model utilizes certified and hardened I/O drivers in the VMware vSphere hypervisor. These drivers must pass rigorous testing and optimization steps performed jointly by VMware and the hardware vendors before they are certified for use with VMware vSphere Hypervisor. With the drivers in the hypervisor, VMware vSphere Hypervisor can provide them with special treatment, in the form of CPU scheduling and memory resources, that they need to process I/O loads from multiple virtual machines. Conversely, the Xen and Microsoft architectures rely on routing all virtual machine I/O to generic drivers installed in the Linux or Windows OS in the hypervisor’s management partition. These generic drivers can be overtaxed by the activity of multiple virtual machines. Windows Server 2008 R2 with Hyper-V and Xen-based products both use generic drivers that are not optimized for many virtual machine workloads that are running concurrently.
VMware investigated the indirect driver model, now used by Xen and Windows Server 2008 with Hyper-V, in early versions of VMware ESX and quickly found that the direct driver model provides much better scalability and performance as the number of virtual machines on a host increases.
Better Memory Management for Scalability
In most virtualization scenarios, system memory is the limiting factor controlling the number of virtual machines that can be consolidated onto a single server. By more intelligently managing virtual machine memory use, VMware vSphere Hypervisor can support more virtual machines on the same hardware than any other x86 hypervisor. Of all x86 bare-metal hypervisors, VMware vSphere Hypervisor supports the broadest set of memory overcommit technologies with minimal performance impact by combining several exclusive technologies.
Content-based transparent memory page sharing conserves memory across virtual machines with similar guest OSs by seeking out memory pages that are identical across the multiple virtual machines and consolidating them so they are stored only once and shared across multiple machines. Think of it as de-duplication for memory. Depending on the similarity of OSs and workloads running on a VMware vSphere Hypervisor host, transparent page sharing alone can typically save anywhere from 5 to 30 percent of the server’s total memory by consolidating identical memory pages (as high as 45% memory savings in a VDI environment). VMware’s balloon driver can also reclaim idle memory so other virtual machines can make use of it. Last, a new memory management technology was added in vSphere 4.1 called memory compression. Compressed memory is a new level of the memory hierarchy, between RAM and disk. Slower than memory, but much faster than disk, this feature improves the performance of virtual machines when memory is under contention, because less virtual memory is swapped to disk as a result.
If all virtual machines on a host spike at the same time and require all of their memory allocation, VMware DRS can automatically load balance by performing VMotion live migrations of virtual machines to other hosts in the DRS cluster.
Watch a technical video on: VMware Distributed Resource Scheduler and VMware vSphere
After dismissing the value of these VMware memory management techniques for quite some time, both Citrix and Microsoft have turned around and released advanced memory capabilities of their own. But in both cases, they still fall far short of the full breadth of what VMware delivers. For instance, Citrix’s dynamic memory in XenServer 5.6 cannot dynamically adjust the size of its balloon driver based on real-time VM usage. This limitation can create major performance issues for active virtual machines while leaving lots of unused RAM in idle virtual machines.
