Abstract
Best-effort hardware transactional memory (HTM) allows complex operations to execute atomically and in parallel, so long as hardware buffers do not overflow, and conflicts are not encountered with concurrent operations. We describe a programming technique and compiler support to reduce both overflow and conflict rates by partitioning common operations into read-mostly (planning) and write-mostly (completion) operations, which then execute separately. The completion operation remains transactional; planning can often occur in ordinary code. High-level (semantic) atomicity for the overall operation is ensured by passing an application-specific validator object between planning and completion. Transparent composition of partitioned operations is made possible through fully-automated compiler support, which migrates all planning operations out of the parent transaction while respecting all program data flow and dependences. For both micro- and macro-benchmarks, experiments on IBM z-Series and Intel Haswell machines demonstrate that partitioning can lead to dramatically lower abort rates and higher scalability.
- Y. Afek, H. Avni, and N. Shavit. Towards consistency oblivious programming. In Proc. of the 15th Intl. Conf. on Principles of Distributed Systems (OPODIS), Toulouse, France, Dec. 2011. Google Scholar
Digital Library
- H. Avni and B. Kuszmaul. Improve HTM scaling with consistency-oblivious programming. In 9th SIGPLAN Wkshp. on Transactional Computing (TRANSACT), Salt Lake City, UT, Mar. 2014.Google Scholar
- H. Avni and A. Suissa. TM-pure in GCC compiler allows consistency oblivious composition. In Joint Euro-TM/MEDIAN Wkshp. on Dependable Multicore and Transactional Memory Systems (DMTM), Vienna Austria, Jan. 2014.Google Scholar
- N. G. Bronson, J. Casper, H. Chafi, and K. Olukotun. Transactional predication: High-performance concurrent sets and maps for STM. In 29th ACM Symp. on Principles of Distributed Computing (PODC), Zurich, Switzerland, July 2010. Google Scholar
Digital Library
- H. W. Cain, M. M. Michael, B. Frey, C. May, D. Williams, and H. Le. Robust architectural support for transactional memory in the Power architecture. In 40th Intl. Symp. on Computer Architecture (ISCA), Tel Aviv, Israel, June 2013. Google Scholar
Digital Library
- C. Click Jr. And now some hardware transactional memory comments. Author’s Blog, Azul Systems, Feb. 2009. www.azulsystems.com/blog/cliff/2009- 02-25-and-now-some-hardware-transactional-memory-comments.Google Scholar
- D. Dice, Y. Lev, M. Moir, and D. Nussbaum. Early experience with a commercial hardware transactional memory implementation. In 14th Intl. Conf. on Architectural Support for Programming Languages and Operating Systems (ASPLOS), Washington, DC, Mar. 2009. Google Scholar
Digital Library
- D. Dice, O. Shalev, and N. Shavit. Transactional Locking II. In 20th Intl. Conf. on Distributed Computing (DISC), Stockholm, Sweden, Sept. 2006. Google Scholar
Digital Library
- P. Felber, V. Gramoli, and R. Guerraoui. Elastic transactions. In 23rd Intl. Conf. on Distributed Computing (DISC), Elche/Elx, Spain, Sept. 2009. Google Scholar
Digital Library
- G. Golan-Gueta, G. Ramalingam, M. Sagiv, and E. Yahav. Concurrent libraries with Foresight. In 34th SIGPLAN Conf. on Programming Language Design and Implementation (PLDI), Seattle, WA, June 2013. Google Scholar
Digital Library
- M. Herlihy and E. Koskinen. Transactional boosting: a methodology for highlyconcurrent transactional objects. In 13th SIGPLAN Symp. on Principles and Practice of Parallel Programming (PPoPP), Salt Lake City, UT, Feb. 2008. Google Scholar
Digital Library
- M. Herlihy, V. Luchangco, M. Moir, and W. N. Scherer, III. Software transactional memory for dynamic-sized data structures. In 22nd ACM Symp. on Principles of Distributed Computing (PODC), Boston, MA, July 2003. Google Scholar
Digital Library
- M. Herlihy and J. E. B. Moss. Transactional memory: Architectural support for lock-free data structures. In 20th Intl. Symp. on Computer Architecture (ISCA), San Diego, CA, May 1993. Google Scholar
Digital Library
- C. Jacobi, T. Slegel, and D. Greiner. Transactional memory architecture and implementation for IBM System z. In 45th Intl. Symp. on Microarchitecture (MICRO), Vancouver, BC, Canada, Dec. 2012. Google Scholar
Digital Library
- C. Lattner and V. Adve. Automatic pool allocation: Improving performance by controlling data structure layout in the heap. In 26th SIGPLAN Conf. on Programming Language Design and Implementation (PLDI), Chicago, IL, June 2005. Google Scholar
Digital Library
- Y. Lev and J.-W. Maessen. Split hardware transactions: True nesting of transactions using best-effort hardware transactional memory. In 13th SIGPLAN Symp. on Principles and Practice of Parallel Programming (PPoPP), Salt Lake City, UT, Feb. 2008. Google Scholar
Digital Library
- M. M. Michael and M. L. Scott. Simple, fast, and practical non-blocking and blocking concurrent queue algorithms. In 15th ACM Symp. on Principles of Distributed Computing (PODC), Philadelphia, PA, May 1996. Google Scholar
Digital Library
- J. E. B. Moss and A. L. Hosking. Nested transactional memory: Model and architecture sketches. Science of Computer Programming, 63(2):186–201, Dec. 2006. Google Scholar
Digital Library
- A. Wang, M. Gaudet, P. Wu, J. N. Amaral, M. Ohmacht, C. Barton, R. Silvera, and M. Michael. Evaluation of Blue Gene/Q hardware support for transactional memories. In 21st Intl. Conf. on Parallel Architectures and Compilation Techniques (PACT), Minneapolis, MN, Sept. 2012. Google Scholar
Digital Library
- L. Xiang and M. L. Scott. MSpec: A design pattern for concurrent data structures. In 7th SIGPLAN Wkshp. on Transactional Computing (TRANSACT), New Orleans, LA, Feb. 2012.Google Scholar
- L. Xiang and M. L. Scott. Compiler aided manual speculation for high performance concurrent data structures. In 18th SIGPLAN Symp. on Principles and Practice of Parallel Programming (PPoPP), Shenzhen, China, Feb. 2013. Google Scholar
Digital Library
- L. Xiang and M. L. Scott. Composable partitioned transactions. In 5th Wkshp. on the Theory of Transactional Memory (WTTM), Jerusalem, Israel, Oct. 2013.Google Scholar
- R. M. Yoo, C. J. Hughes, K. Lai, and R. Rajwar. Performance evaluation of Intel transactional synchronization extensions for high-performance computing. In Intl. Conf. on High Performance Computing, Networking, Storage and Analysis (SC13), Denver, CO, Nov. 2013. Google Scholar
Digital Library
Index Terms
Software partitioning of hardware transactions
Recommendations
Software partitioning of hardware transactions
PPoPP 2015: Proceedings of the 20th ACM SIGPLAN Symposium on Principles and Practice of Parallel ProgrammingBest-effort hardware transactional memory (HTM) allows complex operations to execute atomically and in parallel, so long as hardware buffers do not overflow, and conflicts are not encountered with concurrent operations. We describe a programming ...
Hybrid STM/HTM for nested transactions on OpenJDK
OOPSLA '16Transactional memory (TM) has long been advocated as a promising pathway to more automated concurrency control for scaling concurrent programs running on parallel hardware. Software TM (STM) has the benefit of being able to run general transactional ...
Hybrid STM/HTM for nested transactions on OpenJDK
OOPSLA 2016: Proceedings of the 2016 ACM SIGPLAN International Conference on Object-Oriented Programming, Systems, Languages, and ApplicationsTransactional memory (TM) has long been advocated as a promising pathway to more automated concurrency control for scaling concurrent programs running on parallel hardware. Software TM (STM) has the benefit of being able to run general transactional ...






Comments