A comparison of workload traces from two production parallel machines

Windisch, Kurt; Lo, Virginia; Moore, Reagan; Feitelson, Dror G.; Nitzberg, Bill

doi:10.1109/fmpc.1996.558107

Cited by 36 publications

(17 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These constraints are consistent across typical parallel supercomputing workloads. (For further discussion of parallel supercomputing workloads, see [12]. )…”

mentioning

confidence: 99%

Scheduling for Parallel Supercomputing: A Historical Perspective of Achievable Utilization

Jones

Nitzberg

1999

Job Scheduling Strategies for Parallel Processing

View full text Add to dashboard Cite

Abstract.The NAS facility has operated parallel supercomputers for the past 11 years, including the Intel iPSC/860, Intel Paragon, Thinking Machines CM-5, IBM SP-2, and Cray Origin 2000. Across this wide variety of machine architectures, across a span of 10 years, across a large number of different users, and through thousands of minor configuration and policy changes, the utilization of these machines shows three general trends: (1) scheduling using a naive FCFS first-fit policy results in 40-60% utilization, (2) switching to the more sophisticated dynamic backfilling scheduling algorithm improves utilization by about 15 percentage points (yielding about 70% utilization), and (3) reducing the maximum allowable job size further increases utilization. Most surprising is the consistency of these trends. Over the lifetime of the NAS parallel systems, we made hundreds, perhaps thousands, of small changes to hardware, software, and policy, yet utilization was affected little. In particular, these results show that the goal of achieving near 100% utilization while supporting a real parallel supercomputing workload is unrealistic.

show abstract

“…These constraints are consistent across typical parallel supercomputing workloads. (For further discussion of parallel supercomputing workloads, see [12]. )…”

mentioning

confidence: 99%

Scheduling for Parallel Supercomputing: A Historical Perspective of Achievable Utilization

Jones

Nitzberg

1999

Job Scheduling Strategies for Parallel Processing

View full text Add to dashboard Cite

show abstract

“…The traced job stream is taken only from the 352 nodes [7,9]. The workload trace had the following statistical characteristics: the mean interarrival time was 1186.7 seconds; the average job size was 34.5 nodes, and with the distribution favouring sizes that are non-powers of two.…”

Section: Simulation Resultsmentioning

confidence: 99%

The effect of real workloads and stochastic workloads on the performance of allocation and scheduling algorithms in 2D mesh multicomputers

Bani-Mohammad

Ould‐Khaoua

Mackenzie

et al. 2008

2008 IEEE International Symposium on Parallel and Distributed Processing

View full text Add to dashboard Cite

“…The first is that a relatively large fraction of the jobs are serial (or at least use a single node, on machines such as the SDSC Blue Horizon in which the minimal allocation is an SMP node with 8 processors). The other is that very many jobs use power-of-two nodes [733]. Lublin therefore suggests the following algorithm for selecting a job size [453] (Figure 9.40).…”

Section: Job Sizesmentioning

confidence: 99%

Workload Modeling for Computer Systems Performance Evaluation

Feitelson

2015

196

131

View full text Add to dashboard Cite

Reliable performance evaluations require the use of representative workloads. This is no easy task since modern computer systems and their workloads are complex, with many interrelated attributes and complicated structures. Experts often use sophisticated mathematics to analyze and describe workload models, making these models difficult for practitioners to grasp. This book aims to close this gap by emphasizing the intuition and the reasoning behind the definitions and derivations related to the workload models. It provides numerous examples from real production systems, with hundreds of graphs. Using this book, readers will be able to analyze collected workload data and clean it if necessary, derive statistical models that include skewed marginal distributions and correlations, and consider the need for generative models and feedback from the system. The descriptive statistics techniques covered are also useful for other domains.

show abstract

A comparison of workload traces from two production parallel machines

Cited by 36 publications

References 6 publications

Scheduling for Parallel Supercomputing: A Historical Perspective of Achievable Utilization

Scheduling for Parallel Supercomputing: A Historical Perspective of Achievable Utilization

The effect of real workloads and stochastic workloads on the performance of allocation and scheduling algorithms in 2D mesh multicomputers

Workload Modeling for Computer Systems Performance Evaluation

Contact Info

Product

Resources

About