UPDATED 09:15 EDT / AUGUST 17 2012

OpenStack Isn’t Quite Enterprise-Ready, Says HP Cloud Services Performance Test

OpenStack logoIn a presentation to the OpenStack APAC Conference in Shanghai earlier in August, Qingye Jiang, founder and chief architect of nascent public cloud infrastructure-as-a-service provider ezCloud, describes the results of his stress tests of the OpenStack-powered HP Cloud Services platform. His conclusion: The OpenStack open source cloud platform still has some ways to go ahead of full enterprise readiness.

For his test, Qingye says ezCloud used Ubuntu 11.04 64-bit Server Edition, with three instances of every test VM for comparison. The test, according to Qingye’s blog entry:

“In this test, we used byte-unixbench to evaluate overall system performance, mbw to evaluate memory performance, iozone to evaluate disk IO performance, iperf to evaluate networking performance, pgbench to evaluate database performance, and Hadoop wordcount to evaluate the performance of a complex application.”

For the specifics of the benchmarking tests, the blog entry has the details. Some of the tidbits indicated by the testing includes the interesting possibility that HP provides better disc I/O rates for “better-configured VMs” – which is not something that’s mentioned in the HP Cloud Services documentation. Similarly, larger VMs get better internal bandwidth, it seems.

But the really interesting statistic here is that whenever ezCloud tried provisioning a VM, there was a 35 percent failure rate, which, as Qingye notes, is simply unacceptable for an enterprise offering. Qingye concludes that HP’s internal monitoring is below the par. This is especially dangerous when provisioning VMs automatically by way of API, since you may never realize how many corrupted machines you’re creating.

Qingye says that the fact that OpenStack doesn’t include this kind of system-level monitoring and leaves it to the provider to develop is a sign of platform immaturity:

The defects we encountered in our testings are obviously not introduced by OpenStack itself. However, it implies that OpenStack still lacks certain criticle functionalities such as system level monitoring and notification. Because of the lack of such functionalities, the HP Cloud Services team is not aware of the serious defects that are presented in their system. Therefore, I insist that OpenStack is still not mature enough for commercial deployments.

But on the other hand, OpenStack is itself not much more than a kernel by design, and the community seems to be willing to leave monitoring to the individual community member to develop and refine (and potentially give back into the OpenStack fold), given that different requirements and use-cases necessitate a different approach to the problem.

All the same, there are those who want and need something both open standards-based  and fully-featured. And so, something Qingye and I agree on is that services and support are going to be a major part of OpenStack’s future, for this very reason. Companies like Nebula, Piston and even Rackspace Hosting itself are wrapping custom technology and support offerings around the off-the-shelf OpenStack cloud platform, and that’s going to make all the difference.

 

 

 


A message from John Furrier, co-founder of SiliconANGLE:

Your vote of support is important to us and it helps us keep the content FREE.

One click below supports our mission to provide free, deep, and relevant content.  

Join our community on YouTube

Join the community that includes more than 15,000 #CubeAlumni experts, including Amazon.com CEO Andy Jassy, Dell Technologies founder and CEO Michael Dell, Intel CEO Pat Gelsinger, and many more luminaries and experts.

“TheCUBE is an important partner to the industry. You guys really are a part of our events and we really appreciate you coming and I know people appreciate the content you create as well” – Andy Jassy

THANK YOU