{"id":2372,"date":"2011-06-07T12:00:33","date_gmt":"2011-06-07T11:00:33","guid":{"rendered":"http:\/\/www.markwilson.co.uk\/blog\/?p=2372"},"modified":"2012-09-18T16:00:38","modified_gmt":"2012-09-18T15:00:38","slug":"designing-a-private-cloud-infrastructure","status":"publish","type":"post","link":"https:\/\/www.markwilson.co.uk\/blog\/2011\/06\/designing-a-private-cloud-infrastructure.htm","title":{"rendered":"Designing a private cloud infrastructure"},"content":{"rendered":"

A couple of months ago, Facebook released a whole load of information about its servers and datacentres in a programme it calls the Open Compute Project<\/a>. At around about the same time, I was sitting in a presentation at Microsoft, where I was introduced to some of the concepts behind their datacentres.\u00c2\u00a0 These are not small operations – Facebook’s platform currently serves around 600 million users and Microsoft’s various cloud properties account for a good chunk of the Internet, with\u00c2\u00a0the Windows Azure appliance<\/a> concept under development for partners including Dell, HP, Fujitsu and eBay.<\/p>\n

It’s been a few years since I was involved in any datacentre operations and it’s interesting to hear how times have changed.\u00c2\u00a0Whereas I knew\u00c2\u00a0about redundant uninterruptible power sources and\u00c2\u00a0rack-optimised servers, the model is now about containers of redundant servers and the unit of scale has shifted.\u00c2\u00a0 An appliance used to be a 1U (pizza box) server with a dedicated purpose but\u00c2\u00a0these days it’s a shipping container full of equipment!<\/p>\n

There’s also been a shift from keeping the lights on at all costs, towards efficiency. Hardly surprising, given that the IT industry now accounts for around 3% of the world’s carbon emissions and we need to reduce the environmental impact.\u00c2\u00a0 Google’s datacentre design best practices<\/a> are all concerned with efficiency: measuring power usage effectiveness; measuring managing airflow; running warmer datacentres; using “free” cooling; and optimising power distribution.<\/p>\n

So how do Microsoft (and, presumably others like Amazon too<\/a>) design their datacentres? And how can we learn from them when developing our own private cloud operations?<\/p>\n

Some of the fundamental principles include:<\/p>\n

    \n
  1. Perception of infinite capacity.<\/li>\n
  2. Perception of continuous availability.<\/li>\n
  3. Drive predictability.<\/li>\n
  4. Taking a service provider approach to delivering infrastructure.<\/li>\n
  5. Resilience over redundancy mindset.<\/li>\n
  6. Minimising human involvement.<\/li>\n
  7. Optimising resource usage.<\/li>\n
  8. Incentivising the desired resource consumption behaviour.<\/li>\n<\/ol>\n

    In addition, the following concepts need to be adopted to support the fundamental principles:<\/p>\n