Transparent adaptation of sharing granularity in multiview-based DSM systems

Niv, N.; Schuster, Assaf

doi:10.1109/ipdps.2001.924974

Cited by 4 publications

(6 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nonetheless, our manual analysis can be extended directly to adaptive Millipage by considering different sharing granularities across different iterations and different sections of the iteration. Also, the maximum speedup improvement that is reported in this work [23] ARTICLE IN PRESS [14], and Granston [9] to overcome false sharing effects in software DSM system. While the later two works [9,14] are specifically for fine grain hardware DSM, the first work proposes a software DSM that supports fine grain sharing.…”

Section: Related Workmentioning

confidence: 68%

See 1 more Smart Citation

Performance analysis of methods that overcome false sharing effects in software DSMs

Kudlur

Govindarajan

2004

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

Page-based software DSMs experience high degrees of false sharing especially in irregular applications with fine grain sharing granularity. The overheads due to false sharing is considered to be a dominant factor limiting the performance of software DSMs. Several approaches have been proposed in the literature to reduce/eliminate false sharing. In this paper, we evaluate two of these approaches, viz., the Multiple Writer approach and the emulated fine grain sharing (EmFiGS) approach. Our evaluation strategy is two pronged. First, we use an implementation-independent analysis that uses overhead counts to compare the different approaches. Our analysis show that the benefits gained by eliminating false sharing are far outweighed by the performance penalty incurred due to the reduced exploitation of spatial locality in the EmFiGS approach. As a consequence, any implementation of the EmFiGS approach is likely to perform significantly worse than the Multiple Writer approach. Second, we use experimental evaluation to validate and complement our analysis. The experimental results match well with our analysis. Also the execution times of the application follow the same trend as in our analysis, reinforcing our conclusions. More specifically, the performance of the EmFiGS approach is significantly worse, by a factor of 1.5 to as much as 90 times, compared to the Multiple Writer approach. In many cases, the EmFiGS approach performs worse than even a single writer lazy release protocol which experiences very high overheads due to false sharing.The performance of the EmFiGS approach remains worse than the Multiple Writer approach even after incorporating Tapeworm-a record and replay technique that fetches pages ahead of demand in an aggregated fashion-to alleviate the spatial locality effect. We next present the effect of asynchronous message handling on the performance of different methods. Finally, we investigate the inter-play between spatial locality exploitation and false sharing elimination with varying sharing granularities in the EmFiGS approach and report the tradeoffs.

show abstract

Section: Related Workmentioning

confidence: 68%

“…The successor to Millipage [13,23] extends the technique to adapt the sharing granularity across variables and code sections dynamically. This is a much recent work and is not addressed elaborately in our work.…”

Section: Related Workmentioning

confidence: 99%

Performance analysis of methods that overcome false sharing effects in software DSMs

Kudlur

Govindarajan

2004

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

show abstract

“…A mechanism that changes the memory granularity during runtime can significantly improve MULTIVIEW's performance. An algorithm for modifying the granularity on the fly was discussed in Niv and Schuster [33]. It uses a history of shared memory accesses to aggregate variables to larger coherency units when the application accesses them coarsely.…”

Section: Chunk Allocation and Dynamic Granularitymentioning

confidence: 99%

“…In addition, we try to estimate how the dynamic granularity change can boost the performance of the MULTIVIEW technique. Our system does not support the dynamic granularity protocol, but we try to speculate the runtime performance for the dynamic granularity on the basis of results presented in Niv and Schuster [33]. This speculation is based on the performance gain achieved with dynamic granularity versus fixed granularity performance, and the presented results are only an approximation.…”

Section: Performance Evaluationmentioning

confidence: 99%

“…The aggregation of bodies in chunks does not improve the performance because of strongly increased false sharing. According to Niv and Schuster [33], a dynamic granularity protocol more than doubles SC/MV's speedup, but the resulting speedup still does not come close to HLRC performance. Running two threads per node hardly improves the performance.…”

Section: Performance Analysis Of Barnesmentioning

confidence: 99%

See 1 more Smart Citation

Software Distributed Shared Memory: a VIA-based implementation and comparison of sequential consistency with home-based lazy release consistency

Iosevich

Schuster

2005

Softw: Pract. Exper.

View full text Add to dashboard Cite

A Distributed Shared Memory (DSM) system provides a distributed application with a shared virtual address space. This article proposes a design for implementing the DSM communication layer on top of the Virtual Interface Architecture (VIA), an industry standard for user-level networking protocols on high-speed clusters. User-level communication protocols operate in user mode, thus removing the operating system kernel's overhead from the critical communication pass, and significantly diminishing communication overhead as a result. We analyze VIA's facilities and limitations in order to ascertain which implementation trade-offs can be best applied to our development of an efficient communication substrate optimized for DSM requirements. We then implement a multithreaded version of the Home-based Lazy Release Consistency (HLRC) protocol on top of this substrate. In addition, we compare the performance of this HLRC protocol with that of the Sequential Consistency (SC) protocol in which a MULTIVIEW (MV) memory mapping technique was used. This technique enables a fine-grained access to shared memory, while still relying on the virtual memory hardware to track memory accesses. We perform an 'apple-toapple' comparison on the same testbed environment and benchmark suite, and investigate the effectiveness and scalability of both protocols. ; 35:755-786 SOFTWARE DISTRIBUTED SHARED MEMORY 757 synchronization points are reached; that is, between these synchronization points, the shared memory may appear inconsistent to different processors. These alternate models guarantee, for properlylabeled [4] programs, results equivalent to those of a sequentially consistent system. Informally, a program is properly labeled if the program contains enough synchronization to avoid data races. Synchronization operations are divided into ACQUIRE and RELEASE operations, used respectively to obtain and yield exclusive access to shared data. These operations can be thought of as standard lock operations.Lazy Release Consistency (LRC) [5,6] is a refinement of the Release Consistency (RC) model [4]. The RC model requires that shared memory accesses be performed globally upon a RELEASE operation only. The idea of LRC is to make those accesses visible only to the processor that acquires a lock rather than perform all operations globally. A home-based implementation of LRC (HLRC) was proposed by Iftode [10]. In this implementation each shared page has an assigned home node that always hosts the most updated contents of the page. These updated contents may be fetched by a non-home node that needs an updated version. ContributionThis work compares the runtime performance of two multithreaded memory coherence protocols: a multithreaded implementation of the HLRC model and an efficient multithreaded implementation of the SC model that uses a MULTIVIEW [11,12] memory mapping technique. We also examine and compare the scalability of these protocols to a multithreaded mode of execution. Previous studies proposed non-preemptive multithreading [13] or creati...

show abstract

In-kernel integration of operating system and infiniband functions for high performance computing clusters: a DSM example

Liss

Birk

Schuster

2005

IEEE Trans. Parallel Distrib. Syst.

Self Cite

View full text Add to dashboard Cite

The Infiniband (IB) System Area Network (SAN) enables applications to access hardware directly from user level, reducing the overhead of user-kernel crossings during data transfer. However, distributed applications that exhibit close coupling between network and OS services may benefit from accessing IB from the kernel through IB's native Verbs interface, which permits tight integration of these services. We assess this approach using a sequential-consistency Distributed Shared Memory (DSM) system as an example. We first develop primitives that abstract the low-level communication and kernel details, and efficiently serve the application's communication, memory, and scheduling needs. Next, we combine the primitives to form a kernel DSM protocol. The approach is evaluated using our full-fledged Linux kernel DSM implementation over Infiniband. We show that overheads are reduced substantially, and overall application performance is improved in terms of both absolute execution time and scalability relative to an entirely user level implementation.Index Terms-Hardware/software interfaces, high-speed networks, distributed shared memory, parallel computing.

show abstract

Transparent adaptation of sharing granularity in multiview-based DSM systems

Cited by 4 publications

References 17 publications

Performance analysis of methods that overcome false sharing effects in software DSMs

Performance analysis of methods that overcome false sharing effects in software DSMs

Software Distributed Shared Memory: a VIA-based implementation and comparison of sequential consistency with home-based lazy release consistency

In-kernel integration of operating system and infiniband functions for high performance computing clusters: a DSM example

Contact Info

Product

Resources

About