Modeling of workload in MPPs

Jann, Joefon; Pattnaik, Pratap; Franke, Hubertus; Wang, Fang; Skovira, Joseph; Riordan, Joseph

doi:10.1007/3-540-63574-2_18

Cited by 80 publications

(55 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Osogami and Harchol-Balter show how the first three moments of a distribution can be used to define a Coxian distribution with minimal stages [526] or a combined Erlang-Coxian distribution [527]. Jann et al show how to match the first three moments to a hyper-Erlang distribution [369]. Johnson has proposed a parameterized family of distributions that can be used to match the first four moments of the data by transforming it to the normal distribution (the special case of a simple logarithmic transformation thus induces the lognormal distribution) [382].…”

Section: The Methods Of Momentsmentioning

confidence: 99%

“…However, when the scale parameter of the exponentials is bigger, many more stages are required than when it is small. Specific uses of a hyper-Erlang distribution for workload modeling include matching the first three moments of empirical distributions [369], and creating a distribution with …”

Section: Properties and Usesmentioning

confidence: 99%

“…Jann et al have developed a procedure to match the simplest of these four distributions to given data [369]. This is done in two steps.…”

Section: The Hyper-erlang Distributionmentioning

confidence: 99%

“…For example, Jann et al have used hyper-Erlang distributions to create models that match the first three moments of the modeled data [369] (in practice, differences of 30% between moments calculated from the original data and from samples from the model are common, but the order of magnitude is always correct). But these distributions turn out to be distinctly bimodal, and do not reflect the shape of the original distributions which are much more continuous (Figure 4.1).…”

Section: Variations In Shapementioning

confidence: 99%

“…For example, the Jann model of supercomputer workloads divides the job size scale according to powers of two, and creates an independent model of the runtimes for each range of sizes [369]. These models are completely independent and turn out to be quite different from each other ( Figure 6.30 left).…”

Section: Modeling Correlations With Distributionsmentioning

confidence: 99%

See 4 more Smart Citations

Workload Modeling for Computer Systems Performance Evaluation

Feitelson

2015

191

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

Section: The Methods Of Momentsmentioning

confidence: 99%

Section: Properties and Usesmentioning

confidence: 99%

“…Jann et al have developed a procedure to match the simplest of these four distributions to given data [369]. This is done in two steps.…”

Section: The Hyper-erlang Distributionmentioning

confidence: 99%

Section: Variations In Shapementioning

confidence: 99%

Section: Modeling Correlations With Distributionsmentioning

confidence: 99%

See 3 more Smart Citations

Workload Modeling for Computer Systems Performance Evaluation

Feitelson

2015

191

131

View full text Add to dashboard Cite

show abstract

Workload resampling for performance evaluation of parallel job schedulers

Zakay

Feitelson

2014

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARYEvaluating the performance of a computer system is based on using representative workloads. Common practice is to either use real workload traces to drive simulations or use statistical workload models that are based on such traces. Such models allow various workload attributes to be manipulated, thus providing desirable flexibility, but may lose details of the workload's internal structure. To overcome this, we suggest to combine the benefits of real traces and flexible modeling. Focusing on the problem of evaluating the performance of parallel job schedulers, we partition the trace of submitted jobs into independent subtraces representing different users and then recombine them in various ways, while maintaining features such as long‐range dependence and the daily and weekly cycles of activity. This facilitates the creation of longer workload traces that enable longer simulations, the creation of multiple statistically similar workloads that can be used to gauge confidence intervals, the creation of workloads with different load levels, and increasing the frequency of specific events like large surges of activity. Copyright © 2014 John Wiley & Sons, Ltd.

show abstract

Loosely coordinated coscheduling in the context of other approaches for dynamic job scheduling: a survey

Sodan

2005

Concurrency Computat.: Pract. Exper.

View full text Add to dashboard Cite

SUMMARYLoosely coordinated (implicit/dynamic) coscheduling is a time-sharing approach that originates from network of workstations environments of mixed parallel/serial workloads and limited software support. It is meant to be an easy-to-implement and scalable approach. Considering that the percentage of clusters in parallel computing is increasing and easily portable software is needed, loosely coordinated coscheduling becomes an attractive approach for dedicated machines. Loose coordination offers attractive features as a dynamic approach. Static approaches for local job scheduling assign resources exclusively and non-preemptively. Such approaches still remain beyond the desirable resource utilization and average response times. Conversely, approaches for dynamic scheduling of jobs can preempt resources and/or adapt their allocation. They typically provide better resource utilization and response times. Existing dynamic approaches are full preemption with checkpointing, dynamic adaptation of node/CPU allocation, and time sharing via gang or loosely coordinated coscheduling. This survey presents and compares the different approaches, while particularly focusing on the less well-explored loosely coordinated time sharing. The discussion particularly focuses on the implementation problems, in terms of modification of standard operating systems, the runtime system and the communication libraries.

show abstract

Modeling of workload in MPPs

Cited by 80 publications

References 11 publications

Workload Modeling for Computer Systems Performance Evaluation

Workload Modeling for Computer Systems Performance Evaluation

Workload resampling for performance evaluation of parallel job schedulers

Loosely coordinated coscheduling in the context of other approaches for dynamic job scheduling: a survey

Contact Info

Product

Resources

About