Everything you want to know about Visual Studio ALM and Farming
Brian Harry is a Microsoft Technical Fellow working as the Product Unit Manager for Team Foundation Server. Learn more about Brian.
More videos »
Team Foundation Server is a critical component of any company's assets. It protects the very valuable IP that development teams create. As such it's important that it be reliable and secure. Building TFS on SQL Server went a long way toward accomplishing these goals. SQL provides a robust data platform that includes transactions for data integrity, clustering for high avaibility, mirroring/log shipping for data redundency, online back up for non-impactful disaster recovery, a capable security system and more.
In TFS V1 we leverage many of these features to provide a robust team development system. We continue to make progress on enhacing our story and I wanted to let you know about a new development.
Leveraging SQL's clustering feature we are able to provide a high availablity data tier solution with automated fail-over. Unfortunately, the TFS application tier solution has not been as clean. The solution involves what we have called "warm stand-by". You can install a second (redundant) application tier and, in the event of the primary application tier failing, a set of commands can be run to switch the secondary to primary.
Over the past several months, we've been working with HP to develop a "hot stand-by" solution where the application tier will automatically and transparently fail over in the event of a primary failure. HP has just released a white paper on how to use HP Systems Insight Manager to accomplish this. I'm really happy to be able to add this to the repetoire of TFS administrative and operational options. Between this and SQL clustering you can implement a full, end-to-end redundant system with automatic fail-over.
For more information on administrating and operating a Team Foundation Server check out our documentation here.
Over time, we will continue to improve the TFS operational characteristics. We are looking at features like NLB for a "out of the box" transparent AT fail over solution, leveraging SQL log shipping for "real time" offsite data replication and more.