Building Resilient Cloud Services at Microsoft

How can cloud service providers make sure their services are resilient and reliable? A resilient service should be able to withstand certain issues, remaining fully-functional and therefore being perceived as reliable. But cloud service providers operate in a complex ecosystem, constantly being challenged by reliability-related threats.

This month Asia Futures magazine published an article based on an interview with my colleague David Bills, Microsoft’s chief reliability strategist. In the article, David talks about why resilience is important and provides insight into some of the techniques that can be used to achieve it. Most notably is Resilience Modeling Analysis (RMA), a methodology adapted from the industry-standard technique known as Failure Mode Effects and Analysis (FMEA) to better fit the complexities of cloud services.

I found the article interesting and encourage you to read it and share your thoughts on cloud computing via comments below.

The article is the first in an eight-week series dedicated to resiliency, so you can check back each week with Asia Futures magazine to read the latest.

 

About the Author
Adrienne Hall

General Manager, Trustworthy Computing

Adrienne Hall is a General Manager in the Microsoft Trustworthy Computing group, where she leads a team of information technology (IT) professionals who are focused on the security, privacy, reliability, and accessibility of devices and services built on Microsoft technology. Read more »