VROOM!

VMware® VirtualCenter database stores metadata on the state of a VMware Infrastructure environment and is a key component of VirtualCenter performance.   VirtualCenter 2.5 features a number of enhancements that are aimed at greatly improving the performance and scalability of the VirtualCenter database. This paper presents the performance results of benchmarks we conducted to validate these performance enhancements and to provide best practices information for configuring a  VirtualCenter database. The paper also provides information for sizing the server you use to host the VirtualCenter database based on these performance results. Although the new features in VirtualCenter 2.5 benefit users with any of the supported databases, the examples and performance data presented in this study are specific to Microsoft SQL Server and the paper assumes that you have a working knowledge of SQL Server.

http://www.vmware.com/files/pdf/vc_database_performance.pdf

Our partners at Dell have published the first VMmark results using the new AMD quad-core Barcelona processors. Both the 2-socket (8-core) R805 platform and the 4-socket (16-core) R905 platform have been tested. You can find all of the details on the VMmark results page. If you do the math, you will see that Dell achieved an excellent 1.8x throughput scaling from the 2-socket system to the 4-socket system. Another thing I'd like to point out is that some of the VMmark workloads utilized AMD's Rapid Virtualization Indexing (RVI) technology to improve performance. VMware supports a wide range of virtualization techniques and is able to uniquely leverage both hardware and sofware virtualization technologies in order to provide optimal performance.

Just a quick note to announce that we have released version 1.1 of the VMmark benchmark to our hardware partners. As many of you know, VMmark 1.0 utilized only 32-bit workloads, which was a reasonable mix when the benchmark was first defined roughly three years ago. However, 64-bit applications and OSes are becoming much more prevalent and we need the ability to characterize this more complex reality. To address this, we have transitioned three of the VMmark workloads - Java server, database server, and web server - to 64-bit. In order to maintain comparability with the existing version 1.0 results, we have retained the underlying virtual hardware definitions and load levels for each workload. We need to tie up a few remaining loose ends, but we intend to make VMmark 1.1 generally available very soon. Please stay tuned.