| E. D. Berger, B. G. Zorn, and K. S. McKinley. Reconsidering custom memory allocation. In Proceedings of the Conference on Object-Oriented Programming: Systems, Languages, and Applications (OOPSLA) 2002. |
.... latency, fully pipelined (1 access per cycle) MSHR 32 entries for the L1 caches, 32 entries for L2 Store Buffer 32 entries L1 L2 bus 256 bits wide, 2 cycle latency Memory bus 128 bits wide, 4 cycle latency Physical Memory 512MB, 90 cycle latency, fully pipelined For instance, Berger et al. [8] found that a range of applications spent 0 to over 40 of their execution time performing allocator operations (with an average of 16 ) even though they used optimized custom allocators. based upon the number of instructions required to execute each request on a single threaded superscalar. We ....
....pitfalls of multithreaded memory allocation. Our work demonstrates that this option is indeed simpler for workloads where little interaction is needed between different requests, but suffers a performance penalty. Alternatively, many applications (e.g. Apache and several SPEC benchmarks, see [8]) attempt to improve multithreaded memory allocation by the use of application specific memory allocators. Recent research proposes the use of heap layers to make such allocators more reusable and maintainable [7] Our evaluation of dynamic memory allocation differs from previous work in several ....
E. D. Berger, B. G. Zorn, and K. S. McKinley. Reconsidering custom memory allocation. In Conference on ObjectOriented Programming: Systems, Languages, and Applications, November 2002.
No context found.
E. D. Berger, B. G. Zorn, and K. S. McKinley. Reconsidering custom memory allocation. In Proceedings of the Conference on Object-Oriented Programming: Systems, Languages, and Applications (OOPSLA) 2002.
No context found.
E. D. Berger, B. G. Zorn, and K. S. McKinley. Reconsidering custom memory allocation. In OOPSLA'02 ACM Conference on Object-Oriented Systems, Languages and Applications, ACM SIGPLAN Notices, Seattle, WA, Nov. 2002. ACM Press.
No context found.
E. D. Berger, B. G. Zorn, and K. S. McKinley. Reconsidering Custom Memory Allocation. In Conference on Object-Oriented Programming, Systems, Languages, and Applications, pages 1--12, 2002.
No context found.
E. D. Berger, B. G. Zorn, and K. S. McKinley. Reconsidering custom memory allocation. In OOPSLA'02 ACM Conference on Object-Oriented Systems, Languages and Applications, ACM SIGPLAN Notices, Seattle, WA, Nov. 2002.
No context found.
E. D. Berger, B. G. Zorn, and K. S. McKinley. Reconsidering custom memory allocation. In OOPSLA'02 ACM Conference on Object-Oriented Systems, Languages and Applications, ACM SIGPLAN Notices, Seattle, WA, Nov. 2002. ACM Press.
No context found.
E. D. Berger, B. G. Zorn, and K. S. McKinley. Reconsidering custom memory allocation. In OOPSLA'02 ACM Conference on Object-Oriented Systems, Languages and Applications, ACM SIGPLAN Notices, Seattle, WA, Nov. 2002. ACM Press.
No context found.
Emery D. Berger, Benjamin G. Zorn, and Kathryn S. McKinley. Reconsidering custom memory allocation. In Proceedings of the Conference on Object-Oriented Programming: Systems, Languages, and Applications (OOPSLA) 2002.
No context found.
E. D. Berger, B. G. Zorn, and K. S. McKinley. Reconsidering custom memory allocation. In ACM Conference on Object-Oriented Programming Systems, Languages, and Applications, pages 1--12, Seattle, WA, Nov. 2002.
....for most (but not all) of the applications using custom memory managers that we tested. These results argue for solving memory management problems in a general purpose framework. To address the special needs of server applications, we develop a new memory management abstraction called reaps [13]. Reaps are a hybrid of heaps and regions, which permit only bulk deletion of all objects within separate areas of memory. We verify that programs using regions can achieve significant performance gains over general purpose memory management. We show that our implementation of reaps nearly matches ....
Emery D. Berger, Benjamin G. Zorn, and Kathryn S. McKinley. Reconsidering custom memory allocation. In Proceedings of the Conference on Object-Oriented Programming: Systems, Languages, and Applications (OOPSLA) 2002.
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