Temam O. Investigating optimal local memory performance. Technical Report 97/25, PRiSM, Versailles University, December 1997. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Data Locality Optimizations for Multigrid Methods on Structured.. - Weiß (Correct)
....about future cache references. Thus, the strategy is only of theoretical value. Note, however, that a fully associative cache with optimal replacement strategy will have the minimal number of cache misses any cache of the same size for any possible sequence of references can have [SA93, Tem98] 2.4 State of the Art In the following, an overview of the state of the art in microprocessor technology is given with an emphasis on the cache and memory system. In addition, predecessor models and some already announced, but not yet available, microprocessors are introduced briefly. Readers ....
O. Temam. Investigating Optimal Local Memory Performance. In Proceedings ACM International Conference on Architectural Support for Programming Languages and Operating Systems, San Diego, California, USA, October 1998.
Efficient Profile-Based Evaluation of Randomising Set.. - Vandierendonck, De.. (2001) (1 citation) (Correct)
....a specific reference is a capacity miss. However, we do not pay further attention to the specific choice of capacity misses and treat it as a fixed property of the profiling technique. The references that are labeled as capacity miss are references that are identified by the algorithm described in [19]. 3.2 Computing Scores After the profiles have been gathered the scores are computed. For a set index function H the score is the sum of the counts of the vectors in its null space N(H) Computing this value is straightforward and depends on linear algebra to compute the basis of the null space ....
O. Temam. Investigating optimal local memory performance. In Proceedings of the 8th International Conference on Architectural Support for Programming Languages and Operating Systems, pages 218--227, Nov. 1998.
An Optimal Replacement Algorithm for Balancing.. - Vandierendonck, De.. (Correct)
....since two index functions that simply rename the sets (e.g. the set numbers 0 and 1 are swapped) should be treated equal. IV. DESCRIPTION OF THE OPTIMAL REPLACEMENT ALGORITHM The optimal replacement algorithm for balancing multimodule caches presented here is based on the algorithm of Temam [Tem98] for fully associative caches. This algorithm is centred around the concept of locality edges. When a word is referenced at two times t 1 and t 2 with t 1 t 2 , but not in between, then there is a locality edge from time t 1 to t 2 . This locality edge expresses the reuse of data. Locality ....
....a fully associative cache with n frames, there can be no more than n edges selected at any one time. Temam showed that an algorithm that selects each selectable edge, when the edges are ordered by increasing time of the second reference, is optimal with respect to minimising the cache miss ratio [Tem98] Temam s algorithm can also handle spatial edges that occur when the spatial locality is higher than the temporal locality. Spatial edges are not considered here, since their purpose is to load only those words in a cache block which are effectively used. We only need to handle complete cache ....
[Article contains additional citation context not shown here]
O. Temam. Investigating optimal local memory performance. In Proceedings of the 8th International Conference on Architectural Support for Programming Languages and Operating Systems, pages 218--227, November 1998.
Reducing DRAM Latencies with an Integrated Memory Hierarchy Design - Lin, al. (2001) (14 citations) (Correct)
....extra blocks are expected to be useful. This expectation may be based on profile information [9,25] hardware detection of strided accesses [17] or spatial locality [12,14,25] or compiler annotation of load instructions [23] Optimal off line algorithms for fetching a set of noncontiguous words [24] or a variable sized aligned block [25] on each miss provide bounds on these techniques. Pollution may also be reduced by prefetching into separate buffers [13,23] Our work limits prefetching by prioritizing memory channel usage, reducing bandwidth contention directly and pollution indirectly. ....
Olivier Temam. Investigating optimal local memory performance. In Proceedings of the Eighth Symposium on Architectural Support for Programming Languages and Operating Systems, pages 218--227, October 1998.
Hardware Techniques To Improve The Performance Of The.. - Burger (1998) (10 citations) (Correct)
....and measure a formal upper bound on the effectiveness of caches at reducing communication, and show that the potential exists for up to two orders of magnitude in traffic reduction. This was the first formal bound on cache traffic that we have seen, and it has been extended recently by others [122]. We extend this bound analysis by dissecting the gap between optimal and actual traffic into a breakdown of cache mechanisms, which measures the usefulness of each cache mechanism at reducing memory traffic. Using the results of the bounded traffic analysis, in Chapter 4 we propose a number of ....
Olivier Temam. Investigating Optimal Local Memory Performance. In Proceedings of the Eighth Symposium on Architectural Support for Programming Languages and Operating Systems, pages 218--226, October 1998.
Fast and Accurate Method for Determining a Lower Bound .. - Fursin, O'Boyle.. (2004) Self-citation (Temam) (Correct)
No context found.
Temam O. Investigating optimal local memory performance. Technical Report 97/25, PRiSM, Versailles University, December 1997.
An Overview of Cache Optimization Techniques and Cache-Aware .. - Kowarschik, Weiß (2003) (Correct)
No context found.
O. Temam. Investigating Optimal Local Memory Performance. In Proc. ACM Int. Conference on Architectural Support for Programming Languages and Operating Systems, San Diego, California, USA, 1998.
Analysis of Profiling Information for Cache Sensitive Scheduling - Lindenmaier (1999) (Correct)
No context found.
Olivier Temam, \Investigating Optimal Local Memory Performance", ASPLOS VIII, San Jose, California, Oct. 1998
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