Author: Cynthia Bailey Lee
Publisher:
ISBN:
Category :
Languages : en
Pages : 111
Get Book Here
Book Description
To effectively manage High-Performance Computing (HPC) resources, it is essential to maximize return on the substantial infrastructure investment they entail. One prerequisite to success is the ability of the scheduler and user to productively interact. This work develops criteria for measuring productivity, analyzes several aspects of the user-scheduler relationship via user studies, and develops solutions to some vexing barriers between users and schedulers. The five main contributions of this work are as follows. First, this work quantifies the desires of the user population and represents them as a utility function. This contribution is in four parts: a survey-based study collecting utility data from users of a supercomputer system, augmentation of the Standard Workload Format to enable scheduler research using utility functions, and a model for synthetically generating utility function-augmented workloads. Second, a number of the classic scheduling disciplines are evaluated by their ability to maximize aggregate utility of all users, using the synthetic utility functions. These evaluations show the performance impact of inaccurate runtime estimates, contradicting an oft quoted prior result [55] that inaccuracy of estimates leads to better scheduling. Third, a scheduler optimizing the aggregate utility of all users, using a genetic algorithm heuristic, is demonstrated. This contribution includes two software artifacts: an implementation of the genetic algorithm (GA) scheduler, and a modular, extensible scheduler simulation framework that simulates several classic scheduling disciplines and is interoperable with the Standard Workload Format. Fourth, the ability of users to productively interact with this scheduler by providing an accurate estimate of their resource (run time) needs is examined. This contribution consists of formalizing a frequent casual assertion from the scheduling literature, that users typically "pad" runtime estimates, into an explicit Padding Hypothesis, and then falsifying the hypothesis via a survey-based study of users of a supercomputer system. Specifically, absent an incentive to pad-and including incentives to be accurate-the inaccuracy of runtime estimates only improved from an average of 61% inaccurate to an average of 57% inaccurate. This contribution has implications not only for the proposed genetic algorithm scheduler, but for any scheduler that asks users for an estimate, which currently includes virtually all parallel job schedulers both in production use and proposed in the literature. Fifth, a survey of users of a supercomputer system and associated simulations explore the feasibility of removing one of the defining constraints of the parallel job scheduling problem-the non-preemptability of running jobs. An investigation of users' current checkpointing habits produced a workload labeled with per-job checkpoint information, enabling simulation of a checkpoint-aware GA scheduler that may preempt running jobs as it optimizes aggregate utility. Lifting the non-preemptability constraint improves performance of the GA scheduler by 16% (and 23% compared to classic EASY algorithm), including overhead penalties for job termination and restart.
Author: Art Sedighi
Publisher: Springer
ISBN: 3030145689
Category : Computers
Languages : en
Pages : 132
Get Book Here
Book Description
This book introduces a new scheduler to fairly and efficiently distribute system resources to many users of varying usage patterns compete for them in large shared computing environments. The Rawlsian Fair scheduler developed for this effort is shown to boost performance while reducing delay in high performance computing workloads of certain types including the following four types examined in this book: i. Class A – similar but complementary workloads ii. Class B – similar but steady vs intermittent workloads iii. Class C – Large vs small workloads iv. Class D – Large vs noise-like workloads This new scheduler achieves short-term fairness for small timescale demanding rapid response to varying workloads and usage profiles. Rawlsian Fair scheduler is shown to consistently benefit workload Classes C and D while it only benefits Classes A and B workloads where they become disproportionate as the number of users increases. A simulation framework, dSim, simulates the new Rawlsian Fair scheduling mechanism. The dSim helps achieve instantaneous fairness in High Performance Computing environments, effective utilization of computing resources, and user satisfaction through the Rawlsian Fair scheduler.
Author: Walfredo Cirne
Publisher: Springer
ISBN: 3642358675
Category : Computers
Languages : en
Pages : 281
Get Book Here
Book Description
This book constitutes the thoroughly refereed proceedings of the 16th International Workshop on Job Scheduling Strategies for Parallel Processing, JSSPP 2012, which was held in Shanghai, China, in May 2012. The 14 revised papers presented were carefully reviewed and selected from 24 submissions. The papers cover the following topics: parallel batch scheduling; workload analysis and modeling; resource management system software studies; and Web scheduling.
Author: Faisal Saeed
Publisher: Springer Nature
ISBN: 3030335828
Category : Technology & Engineering
Languages : en
Pages : 1188
Get Book Here
Book Description
This book presents the proceedings of the 4th International Conference of Reliable Information and Communication Technology 2019 (IRICT 2019), which was held in Pulai Springs Resort, Johor, Malaysia, on September 22–23, 2019. Featuring 109 papers, the book covers hot topics such as artificial intelligence and soft computing, data science and big data analytics, internet of things (IoT), intelligent communication systems, advances in information security, advances in information systems and software engineering.
Author: Lucio Grandinetti
Publisher: IOS Press
ISBN: 1586038397
Category : Computers
Languages : en
Pages : 552
Get Book Here
Book Description
Collects in four chapters single monographs related to the fundamental advances in parallel computer systems and their developments from different points of view (from computer scientists, computer manufacturers, end users) and related to the establishment and evolution of grids fundamentals, implementation and deployment.
Author: Michael J. North
Publisher:
ISBN:
Category :
Languages : en
Pages : 448
Get Book Here
Book Description
Author: Udoh, Emmanuel
Publisher: IGI Global
ISBN: 1466620668
Category : Computers
Languages : en
Pages : 393
Get Book Here
Book Description
"This book provides insight into the current trends and emerging issues by investigating grid and cloud evolution, workflow management, and the impact new computing systems have on the education fields as well as the industries"--Provided by publisher.
Author: Michael M. Resch
Publisher: Springer Science & Business Media
ISBN: 3642118518
Category : Computers
Languages : en
Pages : 198
Get Book Here
Book Description
This book covers the results of the 11th and 12th Tera?op Workshop and continued a series initiated by NEC and the HLRS in 2004. As part of the Tera?op Workbench, it has become a meeting platform for scientists, application developers, international experts and hardware designers to discuss the current state and future directions of supercomputing with the aim of achieving the highest sustained application perf- mance. The Tera?op Workbench Project is a collaboration between the High Perf- mance Computing Center Stuttgart (HLRS) and NEC Deutschland GmbH (NEC HPCE) to support users to achieve their research goals using High Performance Computing. The ?rst stage of the Tera?op Workbench project (2004–2008) c- centrated on user’s applications and their optimization for the 72-node NEC SX-8 installation at HLRS. During this stage, numerous individual codes, developed and maintained by researchers or commercial organizations, have been analyzed and - timized. Several of the codes have shown the ability to outreach the TFlop/s thre- old of sustained performance. This created the possibility for new science and a deeper understanding of the underlying physics.
Author: Changjun Hu
Publisher: Springer Nature
ISBN: 9813299878
Category : Computers
Languages : en
Pages : 247
Get Book Here
Book Description
This book constitutes the referred proceedings of two workshops held at the 32nd ACM International Conference on Supercomputing, ACM ICS 2018, in Beijing, China, in June 2018. This volume presents the papers that have been accepted for the following workshops: Second International Workshop on High Performance Computing for Advanced Modeling and Simulation in Nuclear Energy and Environmental Science, HPCMS 2018, and First International Workshop on HPC Supported Data Analytics for Edge Computing, HiDEC 2018. The 20 full papers presented during HPCMS 2018 and HiDEC 2018 were carefully reviewed and selected from numerous submissions. The papers reflect such topics as computing methodologies; parallel algorithms; simulation types and techniques; machine learning.
Author: Geyong Min
Publisher: Springer Science & Business Media
ISBN: 3540498605
Category : Computers
Languages : en
Pages : 1176
Get Book Here
Book Description
This book constitutes the refereed joint proceedings of ten international workshops held in conjunction with the 4th International Symposium on Parallel and Distributed Processing and Applications, ISPA 2006, held in Sorrento, Italy in December 2006. It contains 116 papers that contribute to enlarging the spectrum of the more general topics treated in the ISPA 2006 main conference.
