Category Archives: cloud

OpenStack – The future of private and public clouds?

This week, Rackspace and NASA (a cloud computing pioneer), annouced a major contribution of source code to the open-source community, with the launch of OpenStack – a project to develop the software needed to deploy and operate a fully operational cloud computing solution.

Combining work from Rackspace, who run a large public cloud system, and NASA who were among the first to develop private cloud systems, the new OpenStack system currently consists of “OpenStack Object Store”, a cloud-scale storage solution based on Rackspace Cloud Storage, and the newly developed “OpenStack Compute”, the basis for an Amazon EC2 competitor providing computing infrastructure on demand.

So what do Rackspace get out of this? Well, if things go to plan for Rackspace, then in 5 years you’ll be running your applications on an OpenStack cloud, which Rackspace will manage either in their own data centre as part of the Rackspace public cloud or as a dedicated set of machines in a private cloud they host for you, or even as a hybrid cloud with a baseline cloud computing capacity in an enterprise’s own data centre, with extra capacity available on demand in the Rackspace cloud. Of course, you could choose to work with someone else on OpenStack, but Rackspace will be hoping you choose to stick with a company that obviously knows the code well and has been running it successfully for several years. There’s a video interview with one of the Rackspace Cloud founders on Redmonk where this subject comes up.

While these contributions from Rackspace and NASA are significant pieces of the cloud puzzle, the real work of OpenStack is still to come – they have signed up 25 partner organisations, and are now working hard on completing the development and testing of the systems, and adding functionality.

The possibilities for “OpenStart Compute” in particular are significant, with cooperation from across the industry, we could see the rapid inclusion of technologies like CloudAudit, which helps companies verify the security capabilites of a cloud computing platform, and “Open vSwitch”, a network switch that operates inside the cloud, providing the management and security capabilities of a physical network switch but without many of the limitatioons that go with physical cabling.

Assuming OpenStack develops positively, it’s likely that there will be rapid additions of new systems like an “OpenStack Message Queue”, and “OpenStack Block Storage”, though much of the development will depend on the willingness of contributors to either hand over code that is currently closed source, or to start again with a clean slate and re-develop solutions based on the lessons they’ve previously learnt.

The other possibility is that Amazon continues to take the majority share of the cloud computing market, continues to grow their economies of scale and overall cost leadership, adds functionality to match any new additions to OpenStack (currently Amazon S3 and EC2 more than match OpenStack’s capabilities), and people learn to live with the limitations of a public cloud secutiy model.

Either way, the future of computing is significantly different from the way it operates today for most organisations.

Google offering free "Backups", nothing of the sort

There’s lots of stories floating around the web this weekend about Google offering free backups with their Google Apps product, which on the face of it seems a great additional feature for people worried about data loss.

However, even the quickest read of what Google are offering shows it’s not backup, but online replication, 2 fundamentally different solutions.

Backups protect against the following things amongst others:

  • User error – deleting or overwriting 1 or more files
  • User error – deleting or overwriting entire system including account details, etc
  • Hardware error – hardware loss causes failure/corruption of entire system
  • Hardware error – hardware loss causes deletion/corruption of 1 or more files
  • Software error – software error causes failure/corruption of entire system
  • Software error – software error causes deletion/corruption of 1 or more files

Of these 6 general events that backups protect against, Google’s new online replication service seems to protect against only 2 of them – the hardware error events.

In the event of a user deleting a file or their whole account, Google state that for every action you take “in the background we go and write that data to multiple servers within one data center and also in other data centers”, so if you delete your original file, every other “backup” copy will be deleted too. Same with over-writing it, if your file is protected using revisions in Google Docs then you’ll be ok, but that’s the same offering Google has already provided.

If you delete your account (a fairly stupid move for a user, but something a malicious administrator could do), then you are stuck again – your account deletion will not be protected by these backups.

And finally we get to software errors – we have to look at Google’s statements on the changes again, and according to them the replication is online and instant, so any system caused corruptions or deletions of files in one location will most likely be replicated in the other locations.

That’s not to say what Google is offering isn’t a great addition to the service, it really is, but with these articles you’ll end up with a bunch of mislead users who think their getting something they’re not.

Please, don’t call it a backup if you can’t do a restore!

Cloud computing price comparison stupidity

Microsoft have announced the pricing for their new Azure cloud-computing platform, and there’s been quite a few articles comparing the pricing to that of Amazon’s AWS cloud computing platform, the largest existing cloud provider.

Most have focussed on Microsoft charging 0.5 cents less per hour for a basic Windows instance than Amazon, 12 cents vs 12.5 cents, and whether they’ve done this to start a price war or simply to appear in-line with the existing suppliers out there.

However, these comparisons are just plain stupid, for one reason alone.

Each one provides a completely different definition of a CPU!

Amazon use the “EC2 Compute Unit”, they say that’s based on:

We use several benchmarks and tests to manage the consistency and predictability of the performance of an EC2 Compute Unit. One EC2 Compute Unit provides the equivalent CPU capacity of a 1.0-1.2 GHz 2007 Opteron or 2007 Xeon processor

Microsoft haven’t published a definition of their equivalent CPU definition, but since Amazon haven’t published their exact benchmarks it’s bound to be different.

Again, Google have their App Engine service, where they define their CPU usage as:

CPU time is reported in “seconds,” which is equivalent to the number of CPU cycles that can be performed by a 1.2 GHz Intel x86 processor in that amount of time. The actual number of CPU cycles spent varies greatly depending on conditions internal to App Engine, so this number is adjusted for reporting purposes using this processor as a reference measurement.

The Google measurement is obviously fairly close to the vague Amazon definition of a Compute Unit, but neither of them clearly specify how they actually measure the usage, so any initial comparison is at best vague and at worst completely misleading.

The same is true of Rackspace’s Mosso cloud, and all the other cloud providers out there.

Until a standard CPU unit is defined publicly and agreed between the major suppliers (if that’s even possible), any comparisons between clouds based on a simple “CPU Time” measurement, are simply stupid.