The Product Development Agenda Blog

Amazon's S3 well publicized (Simple Storage Service) experienced another outage over the weekend.  Now I'm a big fan of cloud computing and it's one of the reasons that Symphony is partnering with Salesforce.com to help ISVs adopt a SaaS model utilizing the Force.com platform.  But the very fact that Amazon, who has poured millions into R&D for S3 and EC2 have seen failures should make SaaS ISVs think about

Now most ISVs who are adopting a SaaS model are still philosophically looking at the move as a software engineering challenge.  But the fact is that this SaaS requires a systems engineering perspective.  It's not just the software that gets developed it's how we think about how that software interacts with the underlying infrastructure and how together they deal with internal and external threats -- security vulnerabilities, natural disasters, disaster recovery, etc.  Symphony calls this "Continuity of Operations".

But it is more than a cobbling together amalgamation of various pieces.  Continuity of Operations is more than Disaster Recovery, more than Security.  You need to think in terms of the impact on your business from unplanned outages or loss of customer data.

When you start thinking about continuity of operations, new concepts like Mean-Time-to-Data-Loss (MTTDL) -- the expected amount of time between losing data; think Mean Time Between Failure (MTBF) for your data -- get introduced.  And the difference in terminology is significant.  When most people think about operational metrics they think about availability and scalability...but only in terms of the application.  If the data dissapears, who cares about whether the application is still up.  This has become even more important with regulatory requirements to maintain data for 1, 7, 14 or more years.  SaaS vendors are by definition responsible for their customers' data so the focus on this area is heightened, even more so for companies in the financial and human capital markets. 

How do you think about your SaaS development -- a software engineering enterprise or a systems engineering challenge?  Have you ever heard of MTTDL before you read this article?  Please let me know. 

For more information on the Amazon S3 outage, check out this Network World article.