Scott Drummonds on Virtualization

Server Flash + Array Flash


With the recent general availability of VFCache, EMC has buzzed with ideas about what to do with server-based solid state storage.  Server-based solid state has been around for years.  I remember when Fusion-io visited us at VMware in 2009.  I spent a lot of time thinking about use cases, value, costs, and features. Now at EMC I am asking myself even bigger questions: how far can we go with this technology?  How much can we federate it, migrate data among nodes and within shared storage, protect it, and replicate it?  There are a lot of smart people in EMC that are way ahead of me on this.

But for the time being, the world is using server cache to speed up applications while living with mobility limitations.  Because of my performance background I am still a speed junky.  My long time in that field makes me a bit of a cynic, too.  When I saw a version of the following chart used in an internal EMC presentation I was skeptical.  Take at look at this and ask yourself if you believe it.

Read the rest of this entry »

Virtual Machine Guest OS Optimization

1 Comment »

Last week I was visited in Singapore by Veloxum’s VP of marketing, Chris Cicchetti. Chris was in town presenting Veloxum at a local cloud conference. We both had some extra time and decided to get acquainted over a Thai lunch.

Chris explained that Veloxum works by modifying guest and host (hypervisor) parameters to optimize performance. He claimed that huge gains–frequently 30% and more–could be derived from improving guest OS kernel parameters, application parameters, and host advanced settings.  Veloxum constantly monitors and modifies these parameters to suit changing conditions, which they believe will yield continual benefits in a dynamic environment.

Read the rest of this entry »

Performance Troubleshooting Guide Update


My old friends in VMware’s performance team updated the hugely popular performance troubleshooting guide for vSphere with a vSphere 4.1 version. The previous version of this document has been viewed nearly 30,000 times to date, making it the most popular document ever produced by performance engineering. The vSphere 4.1 version has already been seen over 9,000 times which suggests its early popularity with a large audience.

This paper documents the flow used by VMware’s outbound performance engineering team when they are engaged to investigate a performance problem. The process is shown in flow charts that will lead the reader to a bottleneck. The problems associated with resource bottlenecks are documented in a “cause” and “solution” section towards the end of this publication.

Before I left VMware one of my friends in the VMware support organization told me that the first version of this document was instrumental in educating VMware support on the diagnosis of performance issues. Hal’s document became mandatory reading for anyone that faced a performance problem. With the help of Chethan, his teammate and one of my VMworld co-presenters, this update is surely to add even more value to this document.