Simulating Big Data Clusters for System Planning, Evaluation, and Optimization

Bian, Zhaojuan; Wang, Kebing; Wang, Zhihong; Munce, Gene; Cremer, Illia; Zhou, Wei; Chen, Qian; Xu, Gen

doi:10.1109/icpp.2014.48

Cited by 13 publications

(11 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to the prediction models, the testers can simulate the execution of the programs to analyze their performance in a fine‐grained way. As with the prediction models, the simulators also consider characteristics about the input dataset, the program functionality, the programming cluster, and the file system . The MRPerf simulator considers the inter‐rack and intra‐rack interactions over network using ns‐2 and can be combined with other simulators, such as DiskSim.…”

Section: Resultsmentioning

confidence: 99%

Testing MapReduce programs: A systematic mapping study

Morán

Riva

Tuya

2018

J Software Evolu Process

View full text Add to dashboard Cite

Summary Context MapReduce is a processing model used in Big Data to facilitate the analysis of large data under a distributed architecture. Objective The aim of this study is to identify and categorize the state of the art of software testing in MapReduce applications, determining trends and gaps. Method Systematic mapping study to discuss and classify according to international standards 54 relevant studies in relation to reasons for testing, types of testing, quality characteristics, test activities, tools, roles, processes, test levels, and research validations. Results The principal reasons for testing MapReduce applications are performance issues, potential failures, issues related to the data, or to satisfy the agreements with efficient resources. The efforts are focused on performance and, to a lesser degree, on functionality. Performance testing is carried out through simulation and evaluation, whereas functional testing considers some program characteristics (such as specification and structure). Despite the type of testing, the majority of efforts are focused at the unit and integration test levels of the specific MapReduce functions without considering other parts of the technology stack. Conclusions Researchers have both opportunities and challenges in performance and functional testing, and there is room to improve their research though the use of mature and standard validation methods.

show abstract

Section: Resultsmentioning

confidence: 99%

Testing MapReduce programs: A systematic mapping study

Morán

Riva

Tuya

2018

J Software Evolu Process

View full text Add to dashboard Cite

show abstract

“…The work presented in this paper builds on our previous work, CSMethod [8]. CSMethod enables full-system performance modeling and prediction of big data clusters by simulating both the software stack (e.g.…”

Section: Background and Related Workmentioning

confidence: 99%

“…The hardware model builds on our previous work, CSMethod [8]. Here, we extend CSMethod to enable the modeling of HPC applications.…”

Section: A Hpc Hardware Infrastructure Simulationmentioning

confidence: 99%

See 1 more Smart Citation

Simulation-Based Performance Prediction of HPC Applications: A Case Study of HPL

Xu¹,

Ibeid

Jiang³

et al. 2020

2020 IEEE/ACM International Workshop on HPC User Support Tools (HUST) and Workshop on Programming and Performance Visualization

Self Cite

View full text Add to dashboard Cite

We propose a simulation-based approach for performance modeling of parallel applications on high-performance computing platforms. Our approach enables full-system performance modeling: (1) the hardware platform is represented by an abstract yet high-fidelity model; (2) the computation and communication components are simulated at a functional level, where the simulator allows the use of the components native interface; this results in a (3) fast and accurate simulation of full HPC applications with minimal modifications to the application source code. This hardware/software hybrid modeling methodology allows for low overhead, fast, and accurate exascale simulation and can be easily carried out on a standard client platform (desktop or laptop). We demonstrate the capability and scalability of our approach with High Performance LINPACK (HPL), the benchmark used to rank supercomputers in the TOP500 list. Our results show that our modeling approach can accurately and efficiently predict the performance of HPL at the scale of the TOP500 list supercomputers. For instance, the simulation of HPL on Frontera takes less than five hours with an error rate of four percent.

show abstract

“…Recently, as the demand for big data analysis and neural network processing has increased explosively, the use of data-intensive applications has also drastically increased [1,2]. Because data-intensive processing grows linearly in execution time with data size and has data-level parallelism, its performance can be effectively improved through parallelization [3,4]. Thus, there have been various approaches to accelerate the applications by exploiting data-level parallelism using immense amounts of hardware resources, such as in cloud and neural computing platforms [5][6][7] which are comprised of a variety of state-of-the-art multiple CPUs and GPUs, large-scale memory, high-speed network connections, etc.…”

Section: Introductionmentioning

confidence: 99%

EPSim-C: A Parallel Epoch-Based Cycle-Accurate Microarchitecture Simulator Using Cloud Computing

Kim

Han

2019

Electronics

View full text Add to dashboard Cite

Recently, computing platforms have been being configured on a large scale to satisfy the diverse requirements of emerging applications like big data and graph processing, neural network, speech recognition and so on. In these computing platforms, each computing node consists of a multicore, an accelerator, and a complex memory hierarchy, which are connected to other nodes using a variety of high-performance networks. Up to now, researchers have been using cycle-accurate simulators to evaluate the performance of computer systems in detail. However, the execution of the simulators, which models modern computing architecture for multi-core, multi-node, datacenter, memory hierarchy, new memory, and new interconnection, is too slow and infeasible; since the architecture has become more complex today, the complexity of the simulator is rapidly increasing. Therefore, it is seriously challenging to employ them in the research and development of next-generation computer systems. To solve this problem, we previously presented EPSim (Epoch-based Simulator), which defines epochs that can be run independently by dividing the simulation run into several sections and executes them in parallel on a multicore platform, resulting in only the limited simulation speedup. In this paper, to overcome the computing resource limitations on multi-core platforms, we propose a novel EPSim-C (EPSim on Cloud) simulator that extends EPSim and achieves higher performance using a cloud computing platform. EPSim-C is designed to perform the epoch-based executions in a massively parallel fashion by using MapReduce on Hadoop-based systems. According to our experiments, we have achieved a maximum speed of 87.0× and an average speed of 46.1× using 256 cores. As far as we know, EPSim-C is the only existing way to accelerate the cycle-accurate simulator on cloud platforms; thus, our significant performance enhancement allows researchers to model and research current and future cutting-edge computing platforms using real workloads.

show abstract

Simulating Big Data Clusters for System Planning, Evaluation, and Optimization

Cited by 13 publications

References 9 publications

Testing MapReduce programs: A systematic mapping study

Testing MapReduce programs: A systematic mapping study

Simulation-Based Performance Prediction of HPC Applications: A Case Study of HPL

EPSim-C: A Parallel Epoch-Based Cycle-Accurate Microarchitecture Simulator Using Cloud Computing

Contact Info

Product

Resources

About