Stephen Withers
Tuesday, 16 June 2009 12:47
Business IT -
Technology
Page 2 of 2
If a provider doesn't offer the tools to give that necessary visibility, Foody suggests that you either need to be able to install your own tools on the provider's platform or go elsewhere.
The more enlightened and advanced providers will help, he says, because they know that they are likely to cop the blame for an incident unless they can show their systems are working. Furthermore, transparency builds loyalty, says Foody.
He also suggests that organisations pick providers that have experienced at least one major outage.
This may seem strange advice, but he points out that unless a provider has successfully handled an outage you cannot be confident that the correct and timely steps will be taken when one does occur.
Salesforce.com, for example, has experienced outages, runs its own infrastructure (ie, doesn't rely on other providers), but does not make provision for customers to see the state of the system.
In comparison, Amazon's EC2 service allows you to run any application, including your own diagnostic programs. But Foody admits this is the exception rather than the rule.
It isn't as if there are no models for making chains of services work, says Foody, pointing to examples in the telecommunications, travel and banking industries.