Symbolic expression analysis for compiled communication

Shao, Shuyi; Jones, Alex K.; Melhem, Rami

doi:10.1109/ipdps.2008.4536344

Cited by 2 publications

(2 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To find a communication pattern of the program, its analysis is required. In literature, various techniques such as static analysis, symbolic analysis, and dynamic analysis [3] [4] [5] are proposed for the analysis of MPI program. The data structure DFG (Data Flow Graph) represents the flow of data in the program through different functions and processes.…”

mentioning

confidence: 99%

Data Flow Analysis of MPI Program Using Dynamic Analysis Technique with Partial Execution

Manwade¹,

Kulkarni²

2017

SCPE

View full text Add to dashboard Cite

Message Passing Interface (MPI) is a dominant parallel programming paradigm. MPI processes communicate with each other by sending or receiving messages through communication functions. The application's communication latency will be less if processes are scheduled on nearest cores or nodes and communication latency will be more if processes are scheduled on farthest cores or nodes.The communication latency can be reduced by using topology-aware process placement technique. In this technique, MPI processes are placed on the nearest cores if they have more communication between them. To find the communication pattern between processes, analysis of MPI program is required. Various techniques like static, symbolic and dynamic analysis are available for finding communication pattern of MPI program. These techniques are either taking more time for analysis or fail to find correct communication pattern. In this paper, we have proposed DAPE (Dynamic Analysis with Partial Execution) technique for analysis of MPI program, which finds correct communication pattern in less time as compared to existing techniques. The experimental results show that the proposed technique outperforms over the existing techniques.

show abstract

mentioning

confidence: 99%

Data Flow Analysis of MPI Program Using Dynamic Analysis Technique with Partial Execution

Manwade¹,

Kulkarni²

2017

SCPE

View full text Add to dashboard Cite

show abstract

“…Shao et al [172] present a compiler framework which can identify communication patterns for MPI-based parallel applications. The analysis has been used to minimize power consumption [173], and to guide source code transformations [163]. This approach has the drawback that it is limited for those applications that show static communication patterns.…”

Section: Kim and Liljamentioning

confidence: 99%

Optimizing programming models for massively parallel computers

Farreras Esclusa

View full text Add to dashboard Cite

Since the invention of the transistor, clock frequency increase was the primary method of improving computing performance. As the reach of Moore's law came to an end, however, technology driven performance gains became increasingly harder to achieve, and the research community was forced to come up with innovative system architectures. Today increasing parallelism is the primary method of improving performance: single processors are being replaced by multiprocessor systems and multicore architectures. The challenge faced by computer architects is to increase performance while limited by cost and power consumption. The appearance of cheap and fast interconnection networks has promoted designs based on distributed memory computing. Most modern massively parallel computers, as reflected by the Top 500 list, are clusters of workstations using commodity processors connected by high speed interconnects. Today's massively parallel systems consist of hundreds of thousands of processors. Software technology to program these large systems is still in its infancy. Optimizing communication has become a key to overall system performance. To cope with the increasing burden of communication, the following methods have been explored: (i) Scalability in the messaging system: The messaging system itself needs to scale up to the 100K processor range. (ii) Scalable algorithms reducing communication: As the machine grows in size the amount of communication also increases, and the resulting overhead negatively impacts performance. New programming models and algorithms allow programmers to better exploit locality and reduce communication. (iii) Speed up communication: reducing and hiding communication latency, and improving bandwidth. Following the three items described above, this thesis contributes to the improvement of the communication system (i) by proposing a scalable memory management of the communication system, that guarantees the correct reception of data and control-data, (ii) by proposing a language extension that allows programmers to better exploit data locality to reduce inter-node communication, and (iii) by presenting and evaluating a cache of remote addresses that aims to reduce control-data and exploit the RDMA native network capabilities, resulting in latency reduction and better overlap of communication and computation. Our contributions are analyzed in two different parallel programming models: Message Passing Interface (MPI) and Unified Parallel C (UPC). Many different programing models exist today, and the programmer usually needs to choose one or another depending on the problem and the machine architecture. MPI has been chosen because it is the de facto standard for parallel programming in distributed memory machines. UPC was considered because it constitutes a promising easy-to-use approach to parallelism. Since parallelism is everywhere, programmability is becoming important and languages such as UPC are gaining attention as a potential future of high performance computing. Concerning the communication system, the languages chosen are relevant because, while MPI offers two-sided communication, UPC relays on a one-sided communication model. This difference potentially influences the communication system requirements of the language. These requirements as well as our contributions are analyzed and discussed for both programming models and we state whether they apply to both programming models.

show abstract

Symbolic expression analysis for compiled communication

Abstract: Enabling circuit switching in multiprocessor systems has the potential to achieve more efficient communication with lower cost compared to packet/wormhole switching. However, in

Cited by 2 publications

References 16 publications

Data Flow Analysis of MPI Program Using Dynamic Analysis Technique with Partial Execution

Data Flow Analysis of MPI Program Using Dynamic Analysis Technique with Partial Execution

Optimizing programming models for massively parallel computers

Contact Info

Product

Resources

About