Share This Author
Lowering the Overhead of Nonblocking Software Transactional Memory
This work considers the design of low-overhead, obstruction-free software transactional memory for non-garbage-collected languages and eliminates dynamic allocation of transactional metadata and co-locates data that are separate in other systems, thereby reducing the expected number of cache misses on the common-case code path.
LLAMA: Efficient graph analytics using Large Multiversioned Arrays
The evaluation shows that LLAMA's mutability introduces modest overheads of 3–18% relative to immutable CSR for in-memory execution and that it outperforms state- of-the-art out-of-memory systems in most cases, with a best case improvement of 5x on breadth-first-search.
Adaptive Software Transactional Memory
This paper considers four dimensions of the STM design space and presents a new Adaptive STM (ASTM) system that adjusts to the offered workload, allowing it to match the performance of the best known existing system on every tested workload.
Lowering the Overhead of Software Transactional Memory
A persistent lock-free queue for non-volatile memory
This paper proposes three novel implementations of a concurrent lock-free queue and implemented the various designs and compared their performance overhead to a simple queue design for standard (volatile) memory.
Privatization techniques for software transactional memory
- Michael F. Spear, Virendra J. Marathe, Luke Dalessandro, M. Scott
- Computer SciencePODC '07
- 12 August 2007
It is argued that privatization comprises a pair of symmetric subproblems: private operations may fail to see updates made by transactions that have committed but not yet completed; conversely, transactions that are doomed but have not yet aborted may see Updates made by private code, causing them to perform erroneous, externally visible operations.
An integrated hardware-software approach to flexible transactional memory
- Arrvindh Shriraman, Michael F. Spear, H. Hossain, Virendra J. Marathe, S. Dwarkadas, M. Scott
- Computer ScienceISCA '07
- 9 June 2007
This work describes an alert on update mechanism (AOU) that allows a thread to receive fast, asynchronous notification when previously-identified lines are written by other threads, and a programmable data isolation mechanism (PDI) that allowing athread to hide its speculative writes from otherthreads, ignoring conflicts, until software decides to make them visible.
A comprehensive strategy for contention management in software transactional memory
- Michael F. Spear, Luke Dalessandro, Virendra J. Marathe, M. Scott
- Computer SciencePPoPP '09
- 14 February 2009
Experimental evaluation demonstrates that the overhead of the mechanisms is low, particularly when conflicts are rare, and that the strategy as a whole provides good throughput and fairness, including livelock and starvation freedom, even for challenging workloads.
Lock cohorting: a general technique for designing NUMA locks
This paper presents lock cohorting, a general new technique for designing NUMA-aware locks that is as simple as it is powerful, and allows one to transform any spin-lock algorithm, with minimal non-intrusive changes, into scalable NUma-aware spin-locks.
This article presents lock cohorting, a general new technique for designing NUMA-aware locks that is as simple as it is powerful, and allows one to transform any spin-lock algorithm, with minimal nonintrusive changes, into a scalable NUma-aware spin-locks.