Cloud Predictions Rob Clyde
CLOUD FACT
Cloudbursting – is a concept where when a company/organisation runs out of its computing resources in its internal data centre. The company then “bursts” the additional workload to an external cloud on an on-demand basis. The internal computing resource is the private cloud and the external cloud is typically a public cloud for which the organisation/company gets charged on a pay- per-use basis. When a organisation/company’s deployment has the ability to do cloudbursting or spreading the load to the public cloud, you essentially have a Hybrid Cloud.
Policy-based optimisation also mitigates the need for cloudbursting. If an organisation is short on capacity, they might look at cloudbursting as a way to get more capacity. However, in my experience, I find that most organisations have plenty of capacity, but they just aren’t using it effectively. Typically, I have seen large numbers of servers that are only lightly utilised. Workloads are sprawled haphazardly across the environment, looking more like a fragmented disk drive than a well-optimised system. When new workloads are deployed it is difficult to find an unused server, so new servers are purchased. A vicious cycle occurs when these new servers also become lightly used, making the server sprawl even worse. Instead of bursting to a public cloud for more capacity, organisations can implement policy-based optimisation, which eliminates server sprawl and frees up resources. With enterprise utilisation rates still hovering around 10 to 15%, optimisation can push out the need for capacity-driven cloud bursting, often by years.
IT’S ALL ABOUT THE APPLICATIONS Lastly, I predict that in 2013, the industry will finally re-
alise that private clouds have to be application aware. “
While there is a lot of talk and hype about cloudbursting, I don’t believe it will be significant in 2013
www.cloudcomputingintelligence.com ”
27
Today’s private clouds treat applications as black boxes to be shuffled about and managed in a uniform manner. This is partially driven by the traditional sepa- ration of application experts from operations staff. The application experts request generic resources from operations, but don’t have a way to explicitly state their applications’ specific characteristics. This prevents application-aware optimisation and perpetuates the one-size-fits-none cloud model.
Each application has unique characteristics and successful cloud deployments will plan for that. In order for a cloud to automate application placement and optimisation, it must be aware of the applications’ attributes and constraints. For example, some compo- nents of an application should never be placed on the same physical host, in order to provide high availability. On the other hand, some components should always be placed on the same host, because they need low- latency communication. Cloud management systems must have this level of application awareness in order to manage the cloud optimally. The rapid growth of cloud in 2013 will force this realisation.
LOOKING TO 2013
In summary, I believe 2013 will be a great year for the cloud world. Surely, there will be new technologies, trends, and crises, as there were in 2012. However, not every commonly accepted trend will actually happen. By separating the wheat from the chaff, we can sort through the hype and focus on what matters most. I’d love to hear your thoughts on my predictions; you can email them to me at
cloud@adaptivecomputing.com.
ABOUT THE AUTHOR Robert A. Clyde
Rob Clyde has over 25 years’ experience as an enterprise software executive with demonstrated leadership success at startup companies to large companies, including Symantec and Axent Technologies. An Internet security pioneer and innovator, he is credited with the creation of the first commercial intrusion detection system and led the business unit that developed and brought to market the first security compliance product. As CTO at Symantec, Rob defined Symantec’s technology strategy and was a key part of the management team that drove the company to grow from slightly under $1bn in revenue to over $5bn in five years. Rob is currently a board member and CEO of Adaptive Computing.
January 2013 CCI Magazine
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68