Automatic generation of benchmarks for I/O-intensive parallel applications

Hao, Meng; Zhang, Weizhe; Zhang, You; Snir, Marc; Yang, Laurence T.

doi:10.1016/j.jpdc.2018.10.004

Cited by 13 publications

(10 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…22 Traditional computers often cannot meet the demand for performance when running these applications acquiring and handling massive data. 23 But Cloud platforms have provided a great opportunity to measure the workflow through virtual mirrors. To increase the system utilization of the physical host and the server, disk utilization has been given a high priority.…”

Section: Virtual Machine Consolidationmentioning

confidence: 99%

“…Even in para‐virtualized environments, it is extremely important to figure out the I/O response as the virtualization is handled by the virtualized layer, without the hardware 22 . Traditional computers often cannot meet the demand for performance when running these applications acquiring and handling massive data 23 . But Cloud platforms have provided a great opportunity to measure the workflow through virtual mirrors.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Performance modeling for I/O‐intensive applications on virtual machines

Bhattacharya

Peng

Mao

et al. 2022

Concurrency and Computation

View full text Add to dashboard Cite

Models for virtual machines running on cloud computing systems. Modeling system behaviors of clouds is a grand challenge because the resource utilization in VMs is heterogeneous due to variability in workload conditions. We address this challenging issue by uniquely (1) objectifying the usage prediction of virtualized resources and(2) predicting the performance trends of programs running on clouds. At the heart of the modeling system, we pay particular attention to CPU cores, disk size, main memory space, and input data volume, which serve as important factors for the developed prediction module. We devise two resource-utilization prediction algorithms driven by two distinctive sets of I/O and CPU intensive benchmarks, where one algorithm deals with execution time and the other one revolves around input data size. We investigate the correlation between CPU/disk utilization and VM live migrations. Our system aims at not only providing performance optimization for virtualized resources but also ensuring service level agreement (SLA) and Quality of Service (QoS). The model fits the curve quite well, thereby advocating for the efficiency of the algorithm. The case studies conducted in this project draw the comparisons between the performance of striped and monolithic disks as well as bringing forth the problem of cache coherence that causes hindrance to the experiment. We also deal with the cache-coherence problem to improve the accuracy of our prediction algorithms K E Y W O R D S cache coherence, I/O intensive, linear model, monolithic and split disk, resource utilization INTRODUCTIONThis article is motivated by the rapid growth in demand for computation resources in data centers and the widespread of cloud computing. Modeling system behaviors of clouds is a grand challenge because the resource utilization in VMs is heterogeneous due to variability in workload conditions.We address this challenging issue by uniquely (1) objectifying the usage prediction of virtualized resources and (2) predicting the performance trends of programs running on clouds.This research is inspired by the following four motivations.Motivation 1: virtualization. Cloud computing has changed the landscape of computing resources delivery models. Large-scale virtualized data centers gain popularity thanks to the rapid growth in demand for computational power-driven by modern service applications combined with the shift to the cloud computing model. 1 The virtualization technique emulates physical computers with a mirrored operating system, virtual disks, virtual CPUs, and virtual memory resources. The emergence of virtual machines promises to streamline the on-demand provisioning of software,

show abstract

Section: Virtual Machine Consolidationmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Performance modeling for I/O‐intensive applications on virtual machines

Bhattacharya

Peng

Mao

et al. 2022

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…Users can then run the resulting benchmarks, which exhibit the complex I/O and communication patterns of real-world applications, without having to experience long execution times. Similarly, Hao et al [30] leverage execution traces from real-world parallel applications to automatically generate synthetic MPI programs that mimic the I/O behaviors of these applications without having to execute their computational segments.…”

Section: Related Workmentioning

confidence: 99%

WfChef: Automated Generation of Accurate Scientific Workflow Generators

Coleman

Casanova

Silva

2021

2021 IEEE 17th International Conference on eScience (eScience)

View full text Add to dashboard Cite

The prevalence of scientific workflows with high computational demands calls for their execution on various distributed computing platforms, including large-scale leadershipclass high-performance computing (HPC) clusters. To handle the deployment, monitoring, and optimization of workflow executions, many workflow systems have been developed over the past decade. There is a need for workflow benchmarks that can be used to evaluate the performance of workflow systems on current and future software stacks and hardware platforms.We present a generator of realistic workflow benchmark specifications that can be translated into benchmark code to be executed with current workflow systems. Our approach generates workflow tasks with arbitrary performance characteristics (CPU, memory, and I/O usage) and with realistic task dependency structures based on those seen in production workflows. We present experimental results that show that our approach generates benchmarks that are representative of production workflows, and conduct a case study to demonstrate the use and usefulness of our generated benchmarks to evaluate the performance of workflow systems under different configuration scenarios.

show abstract

“…Trace-based methods: Trace-driven methods, which are frequently used in simulators [8], [7] and benchmark generation tools [24], [25], can capture detailed performance behavior and model the performance of parallel programs automatically. However, trace-driven methods have some limitations.…”

Section: Related Workmentioning

confidence: 99%

Multi-Parameter Performance Modeling Based on Machine Learning with Basic Block Features

Hao

Zhang

Wang

et al. 2019

2019 IEEE Intl Conf on Parallel &Amp; Distributed Processing With Applications, Big Data &Amp; Cloud Computing, Sustainable Com

Self Cite

View full text Add to dashboard Cite

Considering the increasing complexity and scale of HPC architecture and software, the performance modeling of parallel applications on large-scale HPC platforms has become increasingly important. It plays an important role in many areas, such as performance analysis, job management, and resource estimation. In this work, we propose a performance modeling and prediction framework called SmartPred, which utilizes basic block frequencies as features and uses machine learning algorithms to automatically construct multi-parameter performance models with high generalization ability. To reduce the prediction overhead, we propose some feature-filtering strategies to reduce the number of features in the training stage and build a serial program called BBF collector for each target application to quickly collect feature values in the prediction stage. We demonstrate the use of SmartPred on the TianHe-2 supercomputer with six parallel applications. Results show that SmartPred with SVR achieves better prediction than other input parameter-based modeling methods. The average prediction error and average standard deviation of prediction errors of SmartPred are 8.42% and 6.09%, respectively. In the prediction stage, the average prediction overhead of SmartPred is less than 0.13% of the total execution time.

show abstract

Automatic generation of benchmarks for I/O-intensive parallel applications

Cited by 13 publications

References 8 publications

Performance modeling for I/O‐intensive applications on virtual machines

Performance modeling for I/O‐intensive applications on virtual machines

WfChef: Automated Generation of Accurate Scientific Workflow Generators

Multi-Parameter Performance Modeling Based on Machine Learning with Basic Block Features

Contact Info

Product

Resources

About