Transactional memory

Mar 30, 2009 at 3:55 PM
Hi

I work in the BSC Microsoft Research Center where our main focus currently are transactional memories.

Can you tell me if someone from your team did some work in this field? Or maybe someone from the community?


Thanks
Nebojsa Miletic
http://www.bscmsrc.eu/people/nebojsa-miletic
Coordinator
Mar 31, 2009 at 6:05 AM
Hi nmiletic - as a matter of fact a major publication was based around an STM system built on Bartok (not Singularity though). You can read more about it here:

Harris, T., M. Plesko, et al. (2006). Optimizing Memory Transactions. Proceedings of the 2006 ACM SIGPLAN Conference on Programming Language Design and Implementation. Ottawa, Ontario, Canada, ACM: 14-25.
http://research.microsoft.com/en-us/um/people/tharris/papers/2006-pldi.pdf

I'm not aware of any other work. I hope this is helpful. Let me know if you have any other questions.
--
Derrick Coetzee
Microsoft Research Operating Systems Group developer

Mar 31, 2009 at 1:05 PM
Hi Derrick

I am familiar with that paper. But I was asking if there was any research connecting TMs and Singularity because I couldn't find anything myself. In this paper Singularity is mentioned only once (running Cassini on it for testing) and nothing really specific about it was topic of this paper.

I am just looking for some potential new research ideas and this seemed as potentially interesting combination to experiment with. For sure, I will think more about this.


Thanks
--
Nebojsa Miletic
http://www.bscmsrc.eu/people/nebojsa-miletic