发布求助
文献互助 智能选刊 最新文献

2013 National Conference on Parallel Computing Technologies (PARCOMPTECH)最新文献

筛选
英文 中文
Energy aware management framework for HPC systems 高性能计算系统的能源意识管理框架
2013 National Conference on Parallel Computing Technologies (PARCOMPTECH) Pub Date : 2013-10-08 DOI: 10.1109/PARCOMPTECH.2013.6621402
Ankit Kumar, B. Bindhumadhava, N. Parveen
{"title":"Energy aware management framework for HPC systems","authors":"Ankit Kumar, B. Bindhumadhava, N. Parveen","doi":"10.1109/PARCOMPTECH.2013.6621402","DOIUrl":"https://doi.org/10.1109/PARCOMPTECH.2013.6621402","url":null,"abstract":"High Performance Computing (HPC) Systems provide access to high end resources for parallel jobs execution. Resource monitoring and management are the most important aspects of providing a successful HPC environment. Improving performance, reducing energy consumption and operating costs for HPC environment is crucial. There can be different management strategies to manage HPC resources like energy, performance and operating cost based on the overall system's state, the nature of the workload queued and the administrator's choice. As per the current research trends, there is a need to put all these strategies under one umbrella. This paper presents a design of an energy aware framework which bundles all these strategies to autonomically identifying the best suitable resource management strategy. This framework works with the help of multiple intelligent agents and also uses the past knowledge of the application behavior to decide the strategy. We have explained how this framework intends to reduce the energy consumption and operating cost of HPC Systems by selecting the proposed energy management strategy.","PeriodicalId":344858,"journal":{"name":"2013 National Conference on Parallel Computing Technologies (PARCOMPTECH)","volume":"146 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123362654","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 1
Addressing the idiosyncrasies of context-dependent parallelization 解决依赖于上下文的并行化的特性
2013 National Conference on Parallel Computing Technologies (PARCOMPTECH) Pub Date : 2013-10-08 DOI: 10.1109/PARCOMPTECH.2013.6621393
W. Ahmed
{"title":"Addressing the idiosyncrasies of context-dependent parallelization","authors":"W. Ahmed","doi":"10.1109/PARCOMPTECH.2013.6621393","DOIUrl":"https://doi.org/10.1109/PARCOMPTECH.2013.6621393","url":null,"abstract":"Parallelization of sequential code to yield an increase in the execution time on a multiprocessor system is a mature field of research. Patterns or dwarfs that frequently occur among the sequentially coded grand challenge applications have been comprehensively identified and categorized in literature. POV-Ray, a ray tracing based grand challenge application, that also belongs to the SPEC suite of benchmarks, demonstrates a unique as yet unclassified pattern that may place it in a separate class of applications. The execution profile of POV-Ray is exceedingly dependent on an external file supplied as input. Conventional parallelization methods do not consider the impact that an external input file may have on the execution profile of the application. In the case of POV-Ray, a parallelized executable may result in a speedup less than one when used with different input files as a result of the increased communication-to-cost ratio. This paper elaborates on this unique feature of POV-Ray and presents an argument as to why the conventional method of parallelizing will not work in this case. A novel context-dependent parallelization is suggested that uses scaled-down profiling to generate context-dependent executables to ensure a speedup greater than one in all cases.","PeriodicalId":344858,"journal":{"name":"2013 National Conference on Parallel Computing Technologies (PARCOMPTECH)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126669514","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Parallelization of molecular dynamics code 分子动力学代码的并行化
2013 National Conference on Parallel Computing Technologies (PARCOMPTECH) Pub Date : 2013-10-08 DOI: 10.1109/PARCOMPTECH.2013.6621390
Vibhuti Duggal, D. N. V. R. Subrahmanyam, Gaurav Mishra, Kislay Bhatt, R. Kalmady
{"title":"Parallelization of molecular dynamics code","authors":"Vibhuti Duggal, D. N. V. R. Subrahmanyam, Gaurav Mishra, Kislay Bhatt, R. Kalmady","doi":"10.1109/PARCOMPTECH.2013.6621390","DOIUrl":"https://doi.org/10.1109/PARCOMPTECH.2013.6621390","url":null,"abstract":"Different parallel programming paradigms were explored to parallelize the MD-ILAC[1][2][3][4] (Molecular Dynamics for Interaction of Laser with Atomic Clusters). MD-ILAC is a 3 dimensional, relativistic molecular dynamic code to simulate the interaction of intense lasers with atomic clusters. It can simulate the interaction dynamics of various gaseous clusters. Molecular Dynamics approach is especially suited to study this problem, as an atomic cluster is a collection of a few particles. In this paper, we present a comparative study of performance achieved by parallelizing this program using different parallel processing paradigms.","PeriodicalId":344858,"journal":{"name":"2013 National Conference on Parallel Computing Technologies (PARCOMPTECH)","volume":"82 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121366459","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Heuristics aware advance reservation and scheduling (HAARS) mechanism in hybrid (Grid/Cloud) environment 混合(网格/云)环境下启发式感知的提前预约和调度机制
2013 National Conference on Parallel Computing Technologies (PARCOMPTECH) Pub Date : 2013-10-08 DOI: 10.1109/PARCOMPTECH.2013.6621404
Ponsy R. K. Sathia Bama, T. Somasundaram, K. Govindarajan
{"title":"Heuristics aware advance reservation and scheduling (HAARS) mechanism in hybrid (Grid/Cloud) environment","authors":"Ponsy R. K. Sathia Bama, T. Somasundaram, K. Govindarajan","doi":"10.1109/PARCOMPTECH.2013.6621404","DOIUrl":"https://doi.org/10.1109/PARCOMPTECH.2013.6621404","url":null,"abstract":"Grid Resource Broker allocates the user job/application requests to Grid resources based upon the job/application requirements. In some cases, the broker could not be able to run the user application requests due to the non-availability of application execution environment and the required amount of nodes in the single Grid resource. To handle this situation resource broker should have the mechanism to coordinate and allocate the multiple Grid resources called co-allocation. However, the main challenge in the co-allocation mechanism is there is no guarantee in the availability of resources during the application execution that leads to the non-assimilability of the user required Quality of Service (QoS) parameters. In this research work, we have employed the Bipartite-based Heuristics Aware Advanced Reservation and Scheduling (HAARS) mechanism that select and reserve the resources from Grid/Cloud environment in an advance and near optimal manner. The proposed mechanism made use of the open-source software's such as PluS and Haizea for performing advance reservation in the Grid and Cloud environment. The proposed approach guarantees the availability of resources during the application execution, and also it achieves the user required Quality of Service (QoS) requirements.","PeriodicalId":344858,"journal":{"name":"2013 National Conference on Parallel Computing Technologies (PARCOMPTECH)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123408911","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 3
Integrated scheduling and register allocation for multicore architecture 多核体系结构的集成调度和寄存器分配
2013 National Conference on Parallel Computing Technologies (PARCOMPTECH) Pub Date : 2013-10-08 DOI: 10.1109/PARCOMPTECH.2013.6621400
D. Kiran, J. P. Misra, D. Yashas, S. Gurunarayanan
{"title":"Integrated scheduling and register allocation for multicore architecture","authors":"D. Kiran, J. P. Misra, D. Yashas, S. Gurunarayanan","doi":"10.1109/PARCOMPTECH.2013.6621400","DOIUrl":"https://doi.org/10.1109/PARCOMPTECH.2013.6621400","url":null,"abstract":"Multicore architecture has multiple cores tightly integrated on a single die, with each core having private register files. To maximally utilize the processing power of the architecture, a sequential program is split into small parallel regions to run on different cores. Compile time scheduling and register allocation onto each core can be performed in an integrated manner. For such an integrated approach, an algorithm needs not only to schedule the regions of the program effectively but should also have the ability to detect excessive register demands and to reduce register pressure on the fly. In this paper, an algorithm to perform the integrated instruction scheduling and register allocation without affecting the performance is presented and compared with the normal scheduling approaches.","PeriodicalId":344858,"journal":{"name":"2013 National Conference on Parallel Computing Technologies (PARCOMPTECH)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129221752","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 1
Load balancing strategies for SPH SPH的负载均衡策略
2013 National Conference on Parallel Computing Technologies (PARCOMPTECH) Pub Date : 2013-10-08 DOI: 10.1109/PARCOMPTECH.2013.6621394
K. Puri, P. Ramachandran, Pushkar J. Godbole
{"title":"Load balancing strategies for SPH","authors":"K. Puri, P. Ramachandran, Pushkar J. Godbole","doi":"10.1109/PARCOMPTECH.2013.6621394","DOIUrl":"https://doi.org/10.1109/PARCOMPTECH.2013.6621394","url":null,"abstract":"We evaluate the performance of different load balancing algorithms when used with the Smooth Particle Hydro-dynamics (SPH) particle method. We compare several geometric algorithms and a serial graph partitioning algorithm (Metis) in terms of the efficiency and quality of partitioning. We find that the geometric partitioners are better suited with the RIB method producing the best partitions for the problems considered.","PeriodicalId":344858,"journal":{"name":"2013 National Conference on Parallel Computing Technologies (PARCOMPTECH)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129179023","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 2
Performance analysis between aparapi (a parallel API) and JAVA by implementing sobel edge detection Algorithm 通过实现sobel边缘检测算法对并行API和JAVA的性能进行分析
2013 National Conference on Parallel Computing Technologies (PARCOMPTECH) Pub Date : 2013-10-08 DOI: 10.1109/PARCOMPTECH.2013.6621408
Krishan Gopal Gupta, Nisha Agrawal, Samrit Kumar Maity
{"title":"Performance analysis between aparapi (a parallel API) and JAVA by implementing sobel edge detection Algorithm","authors":"Krishan Gopal Gupta, Nisha Agrawal, Samrit Kumar Maity","doi":"10.1109/PARCOMPTECH.2013.6621408","DOIUrl":"https://doi.org/10.1109/PARCOMPTECH.2013.6621408","url":null,"abstract":"This paper presents performance comparison between aparapi (a parallel API for GPU) and java by implementing sobel edge detection Algorithm in java (run on CPU) and aparapi (run on GPU). Our GPU implementation using Aparapi shows speedup of 6x against CPU implementation using java (serial implementation) and speedup of 2x using java prallel implementation (less than 8 threads). Experiments indicate that java threaded version shows speedup up to 4X against Aparapi implementation (more than 8 threads). This comparison study also include implementation of sobel edge detection algorithm on CPU (sequential, threaded version) and aparapi version for enabled on GPU. This article also discusses how to implement Aparapi kernels for data-parallel operations of Typical Edge detection algorithms based on Sobel operator within Java applications The results for performance gains that can be achieved using with and without Aparapi framework.","PeriodicalId":344858,"journal":{"name":"2013 National Conference on Parallel Computing Technologies (PARCOMPTECH)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134361107","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 19
PocketMatch (version 2.0): A parallel algorithm for the detection of structural similarities between protein ligand binding-sites PocketMatch(2.0版):用于检测蛋白质配体结合位点之间结构相似性的并行算法
2013 National Conference on Parallel Computing Technologies (PARCOMPTECH) Pub Date : 2013-10-08 DOI: 10.1109/PARCOMPTECH.2013.6621397
D. Nagarajan, N. Chandra
{"title":"PocketMatch (version 2.0): A parallel algorithm for the detection of structural similarities between protein ligand binding-sites","authors":"D. Nagarajan, N. Chandra","doi":"10.1109/PARCOMPTECH.2013.6621397","DOIUrl":"https://doi.org/10.1109/PARCOMPTECH.2013.6621397","url":null,"abstract":"Knowledge of protein-ligand interactions is essential to understand several biological processes and important for applications ranging from understanding protein function to drug discovery and protein engineering. Here, we describe an algorithm for the comparison of three-dimensional ligand-binding sites in protein structures. A previously described algorithm, PocketMatch (version 1.0) is optimised, expanded, and MPI-enabled for parallel execution. PocketMatch (version 2.0) rapidly quantifies binding-site similarity based on structural descriptors such as residue nature and interatomic distances. Atomic-scale alignments may also be obtained from amino acid residue pairings generated. It allows an end-user to compute database-wide, all-to-all comparisons in a matter of hours. The use of our algorithm on a sample dataset, performance-analysis, and annotated source code is also included.","PeriodicalId":344858,"journal":{"name":"2013 National Conference on Parallel Computing Technologies (PARCOMPTECH)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130078006","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 15
User-priority guided Min-Min scheduling algorithm for load balancing in cloud computing 基于用户优先级的Min-Min云计算负载均衡调度算法
2013 National Conference on Parallel Computing Technologies (PARCOMPTECH) Pub Date : 2013-10-08 DOI: 10.1109/PARCOMPTECH.2013.6621389
Huankai Chen, Frank Z. Wang, N. Helian, Gbola Akanmu
{"title":"User-priority guided Min-Min scheduling algorithm for load balancing in cloud computing","authors":"Huankai Chen, Frank Z. Wang, N. Helian, Gbola Akanmu","doi":"10.1109/PARCOMPTECH.2013.6621389","DOIUrl":"https://doi.org/10.1109/PARCOMPTECH.2013.6621389","url":null,"abstract":"Cloud computing is emerging as a new paradigm of large-scale distributed computing. In order to utilize the power of cloud computing completely, we need an efficient task scheduling algorithm. The traditional Min-Min algorithm is a simple, efficient algorithm that produces a better schedule that minimizes the total completion time of tasks than other algorithms in the literature [7]. However the biggest drawback of it is load imbalanced, which is one of the central issues for cloud providers. In this paper, an improved load balanced algorithm is introduced on the ground of Min-Min algorithm in order to reduce the makespan and increase the resource utilization (LBIMM). At the same time, Cloud providers offer computer resources to users on a pay-per-use base. In order to accommodate the demands of different users, they may offer different levels of quality for services. Then the cost per resource unit depends on the services selected by the user. In return, the user receives guarantees regarding the provided resources. To observe the promised guarantees, user-priority was considered in our proposed PA-LBIMM so that user's demand could be satisfied more completely. At last, the introduced algorithm is simulated using Matlab toolbox. The simulation results show that the improved algorithm can lead to significant performance gain and achieve over 20% improvement on both VIP user satisfaction and resource utilization ratio.","PeriodicalId":344858,"journal":{"name":"2013 National Conference on Parallel Computing Technologies (PARCOMPTECH)","volume":"164 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124585763","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 291
Challenges of monitoring tool for operational indian national grid GARUDA 印度国家电网GARUDA运行监测工具的挑战
2013 National Conference on Parallel Computing Technologies (PARCOMPTECH) Pub Date : 2013-10-08 DOI: 10.1109/PARCOMPTECH.2013.6621396
Karuna Prasad, Himanshu Gupta, N. Mangala, C. Subrata, H. Deepika, Prahlada Rao B.B.
{"title":"Challenges of monitoring tool for operational indian national grid GARUDA","authors":"Karuna Prasad, Himanshu Gupta, N. Mangala, C. Subrata, H. Deepika, Prahlada Rao B.B.","doi":"10.1109/PARCOMPTECH.2013.6621396","DOIUrl":"https://doi.org/10.1109/PARCOMPTECH.2013.6621396","url":null,"abstract":"Computational grid involves the sharing of dynamic and geographically distributed heterogeneous resources to cater to highend scientific problems. The dynamic nature of grid makes monitoring and maintenance of grid health a challenging task. Grid monitoring is a central part of grid as it functions as the dynamic information repository for both system administrators and developers for status checking, troubleshooting, scheduling, performance tuning and analysis. We have developed and deployed a grid-monitoring tool to monitor Indian national grid “GARUDA”. The unique features of it are monitoring network and computational resources, grid middleware, jobs, storage, softwares, and special scientific instruments. It also captures the service degradations of resources and notifies for corrective actions.","PeriodicalId":344858,"journal":{"name":"2013 National Conference on Parallel Computing Technologies (PARCOMPTECH)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116094172","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 2
首页 上一页
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整，为了账户安全，请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们：info@booksci.cn Book学术提供免费学术资源搜索服务，方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号：481959085
Book学术官方微信