LogGP in theory and practice – An in-depth analysis of modern interconnection networks and benchmarking methods for collective operations

Hoefler, Torsten; Schneider, Timo; Lumsdaine, Andrew

doi:10.1016/j.simpat.2009.06.007

Cited by 18 publications

(18 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bell et al [14] directly assess the LogGP values by running three different benchmarks. Hoefler et al [13] have developed a low-overhead approach that directly computes all LogGP parameters from single measurements and without inter-dependencies to other LogGP parameters, avoiding propagation of first-order errors. Moreover, they have also taken into account the fact that most modern communication systems use message-size dependent protocols to optimize communication, so these protocol switches are automatically detected.…”

Section: Assessment Of Logp-based Modelsmentioning

confidence: 99%

“…Using several configuration sets of (N, D, S ), the overhead and gap parameters of the LogGP model can be assessed [13] based on the following expressions:…”

Section: Assessment Of Logp-based Modelsmentioning

confidence: 99%

“…Using this model, the interval detection process is more accurate since the interval boundaries can be detected by two complementary variables simultaneously. This fact leads to a more sensible interval detection than other approaches that use only T g (s), as in [13]. In any case, several LogP-based models can be considered as simplifications of the more complex LoOgGP model.…”

Section: Characterization Of Communication Behaviorsmentioning

confidence: 99%

“…While the main behavioral intervals can be detected during the execution of the PRT T microbenchmarks [13], we propose a post-mortem analysis, considering the overall information of PRT T measurements, that allow us to use data mining techniques. The detection of behavioral changes is performed, once all PRT T values have been measured.…”

Section: Characterization Of Communication Behaviorsmentioning

confidence: 99%

“…The main advantage of the proposed characterization over a manual inspection of the results is that it provides an automatic methodology that facilitates its integration into more general tasks, such as message scheduling or adaptable changes of algorithms [13]. In this case, post-mortem means that the detection process waits for all experimental measurements, and in any case the whole process can be automatically performed.…”

Section: Characterization Of Communication Behaviorsmentioning

confidence: 99%

See 4 more Smart Citations

Automatic parameter assessment of logp-based communication models in MPI environments

Martínez

Blanco

Dominguez

et al. 2010

Procedia Computer Science

View full text Add to dashboard Cite

Communications in modern interconnection networks become a complex issue to deliver the highest performance to the user. In MPI environments, this complexity leads to different communication protocols for different message sizes. The characterization of these different behaviors is important and useful for software developers and network designers. This paper presents an automatic method to obtain a characterization of the communication behavior of a specific MPI environment using LogP-based models. This procedure automatically detects the message sizes where the communication behavior changes due to the influence of different architectural features and protocols. Thereby, the range of message sizes is split in different intervals, in a transparent way to the user. Then, using LogP-based models, each interval is characterized by its own set of parameters. The procedure is based on the parameterized roundtrip-time low-overhead microbenchmark. The detection of different communication behaviors and the assessment of the parameters of the LogP-based models were implemented in an easy to use statistical environment. Real MPI environments were characterized using our proposal.

show abstract

Section: Assessment Of Logp-based Modelsmentioning

confidence: 99%

“…Using several configuration sets of (N, D, S ), the overhead and gap parameters of the LogGP model can be assessed [13] based on the following expressions:…”

Section: Assessment Of Logp-based Modelsmentioning

confidence: 99%

Section: Characterization Of Communication Behaviorsmentioning

confidence: 99%

Section: Characterization Of Communication Behaviorsmentioning

confidence: 99%

Section: Characterization Of Communication Behaviorsmentioning

confidence: 99%

See 3 more Smart Citations

Automatic parameter assessment of logp-based communication models in MPI environments

Martínez

Blanco

Dominguez

et al. 2010

Procedia Computer Science

View full text Add to dashboard Cite

show abstract

Modeling the performance of parallel applications using model selection techniques

Martínez

Blanco

Dominguez

et al. 2013

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARYNowadays, parallel architectures are changing so fast that there is a need for scalable and efficient tools to analyze and predict the performance of parallel applications. Analytical models are proved to be a useful approximation for characterizing parallel algorithms, but developing accurate analytical models is a hard issue, and, in general, they provide coarse performance predictions due to their intrinsic lack of accuracy. In this paper, we describe in detail the Tools for Instrumentation and Analysis (TIA) framework, an easy-touse tool that automatically obtains accurate performance models by means of analytical expressions. This framework automatizes most of its internal tasks, reducing opportunities for human error, and it only requires the user to focus on the metrics and execution parameters that might influence the performance, those that should be considered in the modeling process. Its main advantage over other tools is that TIA uses model selection techniques that allow the automation of the modeling process. As a case of study, the use of TIA to obtain analytical models of different implementations of the broadcast collective communication in a cluster of multicores is shown. The results obtained by TIA are evaluated and compared with theoretical approaches based on the LogGP model.

show abstract

HLog_nGP: A parallel computation model for GPU clusters

Song

Pang

2015

Concurrency and Computation

View full text Add to dashboard Cite

Parallel computation model is an abstraction for the performance characteristics of parallel computers, and should evolve with the development of computational infrastructure. The heterogeneous CPU/Graphics Processing Unit (GPU) systems have been and will be important platforms for scientific computing, which introduces an urgent demand for new parallel computation models targeting this kind of supercomputers. In this research, we propose a parallel computation model called HLog n GP to abstract the computation and communication features of heterogeneous platforms like TH-1A. All the substantial parameters of HLog n GP are in vector form and deal with the new features in GPU clusters. A simplified version HLog 3 GP of the proposed model is mapped to a specific GPU cluster and verified with two typical benchmarks. Experimental results show that HLog 3 GP outperforms the other two evaluated models and can well model the new particularities of GPU clusters. 4881 exascale systems will have much more hardware parallelism, and the development of large applications on them will be confronted with unprecedented challenges [12], including how to effectively reduce communication overhead. Therefore, the parallel computation model for large heterogeneous CPU/GPU clusters should be adept in modeling the communication operations and predicting the communication performance.In the high-performance computing scenario, communication operations can be subdivided into two categories, namely, memory communication and network communication [13]. The former indicates the transmission of data through memory hierarchy while the latter via network links. We herein give a brief survey of the prior work covering the aforementioned categories.Bosque and Pastor [14] proposed the HLogGP model to capture the heterogeneity in both compute nodes and networks. For a heterogeneous systems with M compute nodes, the five parameters of HLogGP are defined in the following forms: latency L and gap per byte G are M M matrices, while overhead o, gap between messages g and computational power P are all M -element vectors. If all the parameters are accurately determined, HLogGP can well predict the communication performance of parallel algorithms. However, HLogGP has two disadvantages: on the one hand, it is difficult to map the O.M 2 / parameters onto specific large-scale heterogeneous systems; on the other hand, it cannot model the memory communication widely witnessed in GPU platforms.With respect to memory communication, Cameron et al. have conducted in-depth study and proposed memory-logP [13] and log n P [6] models. log n P is an extension of memory-logP, and describes the effect of data continuity on communication performance. It consists of five parameters, of which n denotes the total number of communications involved in each message passing process. This model is good at modeling communication relevant to discontinuous data (e.g., Message Passing Interface (MPI) [15] derived data types), yet has routine performance for the continuous data. The r...

show abstract

LogGP in theory and practice – An in-depth analysis of modern interconnection networks and benchmarking methods for collective operations

Cited by 18 publications

References 31 publications

Automatic parameter assessment of logp-based communication models in MPI environments

Automatic parameter assessment of logp-based communication models in MPI environments

Modeling the performance of parallel applications using model selection techniques

HLog_nGP: A parallel computation model for GPU clusters

Contact Info

Product

Resources

About

LogGP in theory and practice – An in-depth analysis of modern interconnection networks and benchmarking methods for collective operations

Cited by 18 publications

References 31 publications

Automatic parameter assessment of logp-based communication models in MPI environments

Automatic parameter assessment of logp-based communication models in MPI environments

Modeling the performance of parallel applications using model selection techniques

HLognGP: A parallel computation model for GPU clusters

Contact Info

Product

Resources

About

HLog_nGP: A parallel computation model for GPU clusters