Windows Server 2008 features a Server Core installation option. Server Core offers a minimal environment for hosting a select set of server roles, including Hyper-V. It features a smaller disk, memory profile, and attack surface. Therefore, we highly recommend that we use the Server Core installation option for Hyper-V servers.
Using Server Core in the root partition leaves additional memory for the VMs. shoul be in mind that additional server roles installed on the server can adversely affect the performance of the virtualization server, especially if they consume significant amounts of CPU, memory, or I/O resources. that means when we use this feature then this application use very low memory of computer. Minimizing the server roles in the root partition is advised, and this has offers additional benefits, such as reducing the attack surface and the frequency of updates.
Minimizing the background activity in idle VMs releases CPU cycles that can be used elsewhere by other VMs or saved to reduce power consumption. when run this application services then use only needed file that means. Windows guests typically use less than 1% of one CPU when they are idle. Here are some best practices for minimizing the background CPU usage of a VM:
1. Install the latest version of VM integration services.
2. Remove the emulated network adapter through the VM settings dialog box (use a synthetic adapter).
3. Remove unused devices such as the CD-ROM and COM port or disconnect their media.
4. Use Windows Server 2008 for the guest operating system.
5. Disable, throttle, or stagger periodic activity, such as backup and defragmentation when appropriate.
6. Review scheduled tasks and services enabled by default.
when we are looking for that last ounce of performance. There are not many performance knobs in Hyper-V which is by design. However if we are looking for that last bit of performance from our Virtual Machines (VM’s) and have already made a good selection for networking and storage you might consider setting the Non-Uniform Access (NUMA) node.
NUMA means each CPU has a different path to memory and those paths can have various lengths. For example CPU 0 on Node 0 when accessing CPU M’s memory on Node X might take 10ns where as CPU 0 on Node 0 accessing CPU N’s memory on Node Y will take 20ns. It’s this difference in memory access times that can impact overall VM performance. The worst case would be for a VM’s Virtual Processor (VP) to be running on a node furthest from where the memory for the VM is allocated.
http://www.palluxo.com/index.php?option=com_content&view=article&id=487:windows-2008-server-brings-better-results-with-less-cpu-usage&catid=90:blogs&Itemid=284
Monday, June 8, 2009
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment