L. Iftode, M. Blumrich, C. Dubnicki, D. Oppenheimer, J. Singh, and K. Li, "Shared Virtual Memory with Automatic Update Support", in Proc. of the  Home/Search   Document Details and Download   Summary   Related Articles   Check

....protocol is still implemented as communicating protocol agents. Several other papers have suggested hardware support for fine grain remote write operations in the network interface [23] 22] One of the recent implementations is the automatic update release consistency (AURC) home based protocol [16]. This implementation is a page based SW DSM which eliminates diffs the compact encoded representation of the differences between the two pages, frequently used in many page based SW DSM systems by using fine grain remote writes for both the application data and the protocol meta data. The ....

L. Iftode, M. Blumrich, C. Dubnicki, D. L. Oppenheimer, J. P. Singh, and K. Li. Shared Virtual Memory with Automatic Update Support. Technical Report TR-575-98, Princeton University, February 1998.

....3 4 Figure 4: Performance of the spectral analysis code on up to four nodes left: speed up and eciency values right: parallelization overhead 6 nisms provided by the SCI VM. 6 Related work Work related to the approach discussed in this work represented by cluster DSM projects like SHRIMP [1, 10] and Cashmere [14] Like the SCI VM discussed in this work, both projects deploy hardware mechanisms to create a hybrid software hardware solution for DSM. In SHRIMP a custom designed network adapter with an automatic global update functionality is used, while Cashmere utilizes DEC s memory ....

L. Iftode, M. Blumrich, C. Dubnicki, D. L. Oppenheimer, J. P. Singh, and K. Li. Shared Virtual Memory with Automatic Update Support. Technical Report TR-575-98, Princeton University, 1998.

.... as well as in optimizing communication layers and recently in structuring applications for software shared memory [80, 67, 21, 44, 65] Protocols and systems have also been developed to exploit the features of network interfaces in system area networks to enhance software shared memory performance [80, 30, 7, 63]. All this research has helped narrow the performance gap between SVM on clusters and hardware DSM systems for an expanding range of applications at the 16 processor scale. Recent studies examine the performance of software shared memory clusters at significant scale (32 to 64 processors) 87, 35, ....

L. Iftode, M. Blumrich, C. Dubnicki, D. L. Oppenheimer, J. P. Singh, and K. Li. Shared virtual memory with automatic update support. In Technical Report TR575 -98, Princeton University, 1998. Also at the 7th Workshop of Scalable Shared Memory Multiprocessors, June 1998.

....sharing by allowing multiple writers [5] Lazy Release Consistency (LRC) 21, 7] further relaxed the RC protocol to reduce protocol overhead. Treadmarks [20] was the rst SVM implementation using the LRC protocol on a network of stock computers. The Automatic Update Release Consistency (AURC) [16] protocol was the rst proposal to take advantage of memory mapped communication to implement an LRC protocol. Home based Lazy Release Consistency (HLRC) 19] proposed a home based approach to improve the performance on large scale machines. Cashmere [22] is an eager Release Consistent (RC) SVM ....

L. Iftode, M. Blumrich, C. Dubnicki, D.L. Oppenheimer, J.P. Singh, and K. Li. Shared Virtual Memory with Automatic Update Support. In Proceedings of the 13th ACM International Conference on SuperComputing, June 1999.

.... and communication architecture) most of the e#orts so far have been in the lower two: relaxed consistency models and protocol implementations to reduce communication frequency and tra#c [4, 21, 47, 2] and additional hardware support in the communication architecture to reduce communication costs [39, 11, 35, 19, 18, 7, 30, 29]. With the relative maturity of protocols, in the last couple of years SVM research has moved to greater emphasis on the application layer and the synergies available across layers. New areas are being emphasized like application driven performance evaluation, application restructuring for SVM ....

....the memory bus and network and relies on the ability to map pages write through in the caches. The fine grain remote writes can be used not only for application data but also for protocol meta data. Results from the implementation show that AURC usually outperforms the all software home based HLRC [18, 7], but not dramatically. The Cashmere system [29] does not rely on snooping but instruments shared writes to generate explicit remote writes as well, using the remote write support in the Memory Channel [17] together with a multiple writer DC protocol and the broadcast support) It achieves ....

L. Iftode, M. Blumrich, C. Dubnicki, D.L. Oppenheimer, J.P. Singh, and K. Li. Shared virtual memory with automatic update support. Technical Report TR-575-98, Princeton, NJ, February 1998.

.... these together, several systems have achieved reasonable performance on clusters with small processor counts [20, 18, 6, 13, 17] Protocols and systems have also been developed to exploit the features of network interfaces in system area networks to enhance software shared memory performance [20, 8, 2, 15]. All this research has helped narrow the performance gap between SVM on clusters and hardware DSM systems for an expanding range of applications at the 16 processor scale. It is now time to examine whether performance can scaleup to larger processor counts, especially on a wide range of ....

L. Iftode, M. Blumrich, C. Dubnicki, D. L. Oppenheimer, J. P. Singh, and K. Li. Shared virtual memory with automatic update support. In Technical Report TR-575-98, Princeton University, 1998. Also at the 7th Workshop of Scalable Shared Memory Multiprocessors, June 1998.

No context found.

L. Iftode, M. Blumrich, C. Dubnicki, D.L. Oppenheimer, J.P. Singh, and K. Li. Shared Virtual Memory with Automatic Update Support. Technical Report TR-575-98, Princeton, NJ, February 1998.

....and SHRIMP [5] a PC based cluster interconnected with a custom designed network interface) The HLRC implementation on Typhoon zero used active messages for communication support. Our HLRC implementation uses the VMMC 2 library [9] and in this sense is similar to the HLRC implementation on SHRIMP [11] which also uses a memory mapped communication library. SVM performance has been studied, both in simulations [3, 13, 14, 17] and on real systems [1, 15, 16] But our work is the first, to our knowledge, in incorporating the SVM system and a performance debugging tool embedded into Myrinet ....

L. Iftode, M. Blumrich, C. Dubnicki, et al. Shared Virtual Memory with Automatic Update Support. Technical Report TR-575-98, Department of Computer Science, Princeton University, 1998.

.... these together, several systems have achieved reasonable performance on clusters with small processor counts [20, 18, 6, 13, 17] Protocols and systems have also been developed to exploit the features of network interfaces in system area networks to enhance software shared memory performance [20, 8, 2, 15]. All this research has helped narrow the performance gap between SVM on clusters and hardware DSM systems for an expanding range of applications at the 16 processor scale. It is now time to examine whether performance can scaleup to larger processor counts, especially on a wide range of realistic ....

L. Iftode, M. Blumrich, C. Dubnicki, D. L. Oppenheimer, J. P. Singh, and K. Li. Shared virtual memory with automatic update support. In Technical Report TR-575-98, Princeton University, 1998. Also at the 7th Workshop of Scalable Shared Memory Multiprocessors, June 1998.

No context found.

L. Iftode, M. Blumrich, C. Dubnicki, D. Oppenheimer, J. Singh, and K. Li, "Shared Virtual Memory with Automatic Update Support", in Proc. of the

No context found.

L. Iftode, M. Blumrich, C. Dubnicki, D. L. Oppenheimer, J. P. Singh, and K. Li. Shared Virtual Memory with Automatic Update Support. Technical Report TR-575-98, Princeton University, February 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