The Virtualization Practice

Tag Archive for OpenStack

One of the most intriguing names that has hitherto been at the periphery of the OpenStack initiative is Citrix. Up until last week, Citrix’s contribution was to ensure OpenStack ran on XenServer. However, this week at it’s Synergy event, Citrix made some more sigificant announcements about Project Olympus, through which it aims to provide (in collaboration with Dell and Rackspace) a route to commercial exploitation of the OpenStack codebase. For some time I have been perplexed as to what Citrix is doing. Are they genuinely intending to enter this space? Is this the real play or is it a spoiler?

Citrix’s annual Synergy conference held this week in San Francisco was kicked off with CEO Mark Templeton painting his view of the future, and the building and leveraging of cloud services. With the emergence and evolution of cloud services, Templeton believes that the industry has moved out of the PC (personal computing) era into a PC-3 era, incorporating personal, private, and public cloud services.

Amazon failed because of simultaneous failure of its EBS in two Availability zones. If you were dependent on one of these (or mirrored across the two) you lost access to the filesystem from your Instances. It may be sensible to move to the use of the S3 mechanism (or some portable abstraction over it) for new applications, but if you have an existing application that expects to see a filesystem in the traditional way, Gluster can provide a distributed cloud-agnostic shared filesystem with multi-way replication (including asynchronous replication).

Amazon’s Service Level Agreement (SLA) is so narrowly-drawn that it could easily be argued that the recent Elastic Block Store (EBS) outage wasn’t a failure of Amazon Web Services at all. Anyone using EBS in a production environment was, arguably, reaping the fruits of their own folly. Of course they don’t tell you when you read the hype that architecting for resilience in the Cloud is actually very complicated, particularly if you want to take the sensible step of not relying on a single provider like Amazon, no-matter how dominant their hype may be.

Facebook (which had previously bought commodity servers and rented data center space) has opened up a whole new area of Open Source technology by publishing the full specification of both its new custom server and its new data center as “Open Source” at OpenCompute.org. Overall, Facebook claims that its new data centers are 38 per cent more efficient than its existing leased data centers, but the cost is about 20 per cent less. Published data (such as it exists) indicates that Facebook is at or ahead of rivals or peers such as Microsoft and Google. OpenCompute designs are released under new set of Open Source agreements. The intent seems to be to allow innovation within the published specification, but to ensure multiple providers of the technology. Facebook is clearly seeking to get multiple tier-1 third-party providers for both servers and data centers according to these designs, turning these Open Source specifications into a form of de-facto Standard, which could have broad impact by driving the marketplace away from shared storage models (such as Red Hat’s IAAS reference architecture) to local-storage-friendly IAAS architectures such as OpenStack or Eucalyptus.

On October 22nd, Microsoft announced that it has partnered with Cloud.com to provide integration and support of Windows Server 2008 R2 Hyper-V to the OpenStack project. The announcement caused a great deal of interest here at the Virtualization Practice, as it signals an unexpected willingness on Microsoft’s part to pursue interoperability at the IaaS layer, allowing users to break out of the Hyper-v stack, whilst still retaining Hyper-v at the bottom. The fact this announcement came from Microsoft (not Cloud.com, Rackspace or OpenStack) seems to signal the seriousness of the intent.

Eucalyptus-based solution that is bundled into the Ubuntu installation from 9.10 onwards and allows you to install a IaaS cloud into which you subsequently install Ubuntu Server instances, rather than directly installing an Ubuntu Server. The Eucalyptus proposition is that the cloud you create is identical from an API – and therefore a tooling – perspective to an Amazon EC2 cloud, and the same Ubuntu instances can run inside it, and even can be cloud-bursted out to it. Canonical make a lot of this duality in their positioning of Eucalyptus and the Ubuntu Enterprise Cloud. It feels very-much like an “onramp” message that we hear from VMware.

Rackspace has got the OpenStack governance model spectacularly wrong, and as a result the whole initiative is in peril. Not only are the Chair and the Chief Architect appointed directly by Rackspace, but 3 additional members are appointed directly by Rackspace, meaning that the 4 independently-elected Community members (even if they could agree) could never form a majority. There is actually no need to gain control explicitly. You control by contribution. Since Rackspace contributes most it will gain most control. Rackspace doesn’t actually need control to satisfy its business objectives. ll it needs is to make sure the project is successful and retain enough control over the project to ensure its own needs are met. So our suggestion to OpenStack is to take their Governance model, rip it up and start again.