An approach for realistically simulating the performance of scientific applications on high performance computing systems

Mohammed, Ali; Eleliemy, Ahmed; Ciorba, Florina M.; Kasielke, Franziska; Banicescu, Ioana

doi:10.1016/j.future.2019.10.007

Cited by 11 publications

(13 citation statements)

References 48 publications

(96 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The present work proposed a resourceful coordination approach (RCA) that allows application schedulers to cooperate by involving the batch scheduler. We implemented the proposed approach in a two-level simulation using realistic and well-known simulators (a Slurm-based simulator [31] and a SimGrid-based simulator [14]). The effective system performance (ESP) benchmark was used to assess the proposed approach by instantiating it with the parallel spin-image generation and the Mandelbrot set.…”

Section: Discussionmentioning

confidence: 99%

“…At application-level scheduling, the present work designs and extends an accurate SimGrid-based simulator [14] that is used to simulate applications' executions with various DLS techniques by simultaneously simulating the execution of several applications running on the same simulated HPC platform. The intention behind this difference is to let the simulator account for application interference.…”

Section: App4mentioning

confidence: 99%

“…The intention behind this difference is to let the simulator account for application interference. Earlier research efforts [33,14] focused on the study of applications' performance under various scheduling techniques. In contrast, the current work relaxes the assumption of applications executing on separate sets of resources during their entire execution, thereby increasing the realism of the simulation.…”

Section: App4mentioning

confidence: 99%

“…To implement RCA a Slurm-based simulator [11] is employed as a batch scheduling framework together with a SimGrid-based simulator [14] as an application scheduling framework. Both SimGrid-based and Slurm-based simulators have been shown to be realistic in terms of the close agreement between results obtained natively via direct experiments on HPC systems and results obtained via simulation [11,14].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Resourceful Coordination Approach for Multilevel Scheduling

Eleliemy,

Ciorba

2021

Preprint

Self Cite

View full text Add to dashboard Cite

HPC users aim to improve their execution times without particular regard for increasing system utilization. On the contrary, HPC operators favor increasing the number of executed applications per time unit and increasing system utilization. This difference in the preferences promotes the following operational model. Applications execute on exclusively-allocated computing resources for a specific time, and applications are assumed to utilize the allocated resources efficiently. In many cases, this operational model is inefficient, i.e., applications may not fully utilize their allocated resources. This inefficiency results in increasing application execution time and decreasing system utilization. In this work, we propose a resourceful coordination approach (RCA) that enables the cooperation between, currently independent, batch-and application-level schedulers. RCA enables application schedulers to share their allocated but idle computing resources with other applications through the batch system. The effective system performance (ESP) benchmark is used to assess the proposed approach. The results show that RCA increased system utilization up to 12.6% and decreased system makespan by the same percent without affecting applications' performance.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: App4mentioning

confidence: 99%

Section: App4mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Resourceful Coordination Approach for Multilevel Scheduling

Eleliemy,

Ciorba

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…In [28] the authors have used load balancing to speed up matrix-vector multiplication in large scale machine learning and data mining applications. The authors of [29] have developed a simulator to select the best balancing technique for scientific computing applications whose stochastic and irregular natures greatly contribute to load imbalance.…”

Section: Related Workmentioning

confidence: 99%

Fine Tuning the Performance of Parallel Codes

Gheibi¹,

Banerjee²,

Ranka³

et al. 2020

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

We propose a multilevel method to speed highly optimized parallel codes whose runtime increases faster than their workload. This method requires the ability to solve large instances by decomposing them into smaller instances. Using a simple parallel computing model, we derive a mathematical model that predicts whether or not our method can improve performance and also predicts the amount of improvement attainable. Our method is tested and shown to be effective on three highly optimized BLAS (Basic Linear Algebra Subprograms) routines from Intel's Math Kernel Library (MKL). Those routines are cblas dgemm, cblas dtrmm and cblas dsymm. On the Intel Knights Landing (KNL) platform our method speeds cblas dgemm by 33%, cblas dtrmm by 50% and cblas dsymm by 49% on double-precision matrices of size 16K × 16K, when the KNL's default memory-clustering configuration (cache-quadrant) is used.

show abstract

SimAS: A simulation‐assisted approach for the scheduling algorithm selection under perturbations

Mohammed

Ciorba

2020

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

Many scientific applications consist of large and computationally-intensive loops, such as N-body, Monte Carlo, and computational fluid dynamics These loops contain computationally-intensive operations, resulting in heavy loop bodies.Dynamic loop self-scheduling (DLS) techniques are used to parallelize and to balance the load during the execution of such applications. Load imbalance arises from variations in the loop iteration (or tasks) execution times, caused by problem, algorithmic, or systemic characteristics. The variations in systemic characteristics are referred to as perturbations, and can be caused by other applications or processes that share the same resources, or a temporary system fault or malfunction and include, decreased delivered computational speed, reduced available network bandwidth, or larger network latencies. DLS achieves a balanced load execution of scientific applications on high-performance computing (HPC) systems. Therefore, the selection of the most efficient DLS technique is critical to achieve the best application performance. The following question motivates this work: "Given an application, an HPC system, and their characteristics and interplay, which DLS technique will achieve improved performance under unpredictable perturbations?" Existing studies focus on variations in the delivered computational speed only as the source of perturbations in the system. However, perturbations in available network bandwidth or latency are inevitable on production HPC systems. Also, scheduling solutions based on optimization techniques, e.g., evolutionary algorithms, can not adapt to perturbations during execution. The alternative of using machine learning for DLS selection requires training and learning either prior to execution or during previous time-steps in time-stepping applications. A Simulator-assisted scheduling (SimAS ) is introduced as a new control-theoretic-inspired approach to dynamically select DLS techniques that improve the performance of applications executing on heterogeneous HPC systems under perturbations. The present work examines the performance of seven applications on a heterogeneous system under all the above system perturbations. SimAS is evaluated as a proof of concept using native and simulative experiments. The performance results confirm the original hypothesis that no single DLS technique can deliver the absolute best performance in all scenarios, whereas the SimAS -based DLS selection resulted in improved application performance in most experiments. SS and AWF-C are the most efficient in pea-es (b) Application performance under various perturbations using different scheduling techniques.designed for time-stepping applications. It improves WF by adapting the relative weights of PEs during execution by monitoring their performance in each time-step. AWF-B relieves the time-stepping requirement in AWF, and measures the performance after each batch to update the PE weights. AWF-C is similar to AWF-B where weight updates are performed upon the completion of each chunk, ...

show abstract

An approach for realistically simulating the performance of scientific applications on high performance computing systems

Cited by 11 publications

References 48 publications

A Resourceful Coordination Approach for Multilevel Scheduling

A Resourceful Coordination Approach for Multilevel Scheduling

Fine Tuning the Performance of Parallel Codes

SimAS: A simulation‐assisted approach for the scheduling algorithm selection under perturbations

Contact Info

Product

Resources

About