D. J. Scales, K. Gharachorloo, and C. A. Thekkath, Shasta: A Low Overhead, Software-Only Approach for Supporting Fine-Grained Shared Memory, In Proceedings of the Seventh International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS'96), October 1996, pp. 174-185. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Processor Mechanisms for Software Shared Memory - Carter (Correct)
....improvements in remote memory access times at a low hardware cost. The sharedmemory protocol implemented for this thesis is able to resolve a remote memory reference in as little as 336 cycles, more than an order of magnitude faster than software only shared memory systems such as SHASTA [32], and less than 2.5x the remote access time of the SGI Origin 2000 [20] a contemporary system which implements its shared memory protocol completely in hardware. Implementing the M Machine s mechanisms for shared memory requires less than 3 KB of RAM for the event system, global translation ....
....to software on references to invalid pages. One major weakness of IVY was the false sharing which resulted from the large grain size of the shared memory system, which could not move blocks of data smaller than the operating system s page size between processors. To address this limitation, Shasta [32] and Blizzard S [33] modify program executables to add checking code before each remote memory reference which determines whether or not the referenced data is present on the node. This approach allows the shared memory system to trade off block granularity against the amount of data required to ....
[Article contains additional citation context not shown here]
Daniel J. Scales, Kourosh Gharachorloo, and Chandramohan A. Thekkath. Shasta: a low overhead, software-only approach for supporting fine-grained shared memory. In Proceedings of the Seventh International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS-VII), pages 174-185, October 1996.
Enhancing Software DSM for Compiler-Parallelized Applications - Keleher, Tseng (1997) (20 citations) (Correct)
....aware of only a few projects combining compilers and software DSMs. Bershad et al. maintain coherence by using a compiler to update a software dirty bit on sharedmemory accesses [2] Scales et al. designed Shasta, a software only approach that supports fine grain coherence through binary rewriting [24]. Using a number of optimizations, Shasta limits software overhead to within 5 35 for the Splash benchmarks on a DEC Alpha cluster. In comparison, CVM, like most software DSMs, relies on the virtual memory system to detect shared memory updates. Results, however, show that this is not a problem ....
D. Scales, K. Gharachorloo, and C. Thekkath. Shasta: A low overhead, software-only approach for supporting finegrained shared memory. In Proceedings of the Eighth InternationalConference on Architectural Support for Programming Languages and Operating Systems (ASPLOS-VIII), Boston, MA, October 1996.
Enhancing Software DSM for Compiler-Parallelized Applications - Pete Keleher (1997) (20 citations) (Correct)
....of only a few projects combining compilers and software DSMs. Bershad et al. maintain coherence by using a compiler to update a software dirty bit on shared memory accesses [1] Scales et al. designed Shasta, a software only approach that supports fine grain coherence through binary rewriting [24]. Using a number of optimizations, Shasta limits software overhead to within 5 35 for the Splash benchmarks on a DEC Alpha cluster. In comparison, CVM, like most software DSMs, relies on the virtual memory system to detect shared memory updates. Results, however, show that this is not a problem ....
D. Scales, K. Gharachorloo, and C. Thekkath. Shasta: A low overhead, software-onlyapproach for supportingfine-grained shared memory. In Proceedings of the Eighth International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS-VIII), Boston, MA, Oct. 1996.
OpenMP Performance Analysis Approach in the INTONE.. - Ayguadé, Brorsson.. (Correct)
No context found.
D. J. Scales, K. Gharachorloo, and C. A. Thekkath, Shasta: A Low Overhead, Software-Only Approach for Supporting Fine-Grained Shared Memory, In Proceedings of the Seventh International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS'96), October 1996, pp. 174-185.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC