EasyGrid Enabling of Iterative Tightly-Coupled Parallel MPI Applications

Sena, Alexandre C.; Nascimento, Ad.P.; Boeres, Cristina; Rebello, Vinod E. F.

doi:10.1109/ispa.2008.122

Cited by 7 publications

(6 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By coupling legacy MPI applications with EasyGrid AMS, they can be transformed into autonomic versions which manage their own execution. The benefits of this approach include adopting (scheduling, communication, fault tolerance) policies tailored to the specific needs of each application thus leading to improved performance [52]. While the EasyGrid AMS is being used to accelerate phases of openModeller through parallelisation, given that workflows can be seen to be directed acyclic graphs, the AMS can also be used to encapsulate the entire workflow and manage their execution in distributed systems without workflow managers.…”

Section: E Computing Servicesmentioning

confidence: 99%

Supporting biodiversity studies with the EUBrazilOpenBio Hybrid Data Infrastructure

Amaral

Badía

Blanquer

et al. 2014

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

Abstract-EUBrazilOpenBio is a collaborative initiative addressing strategic barriers in biodiversity research by integrating open access data and user-friendly tools widely available in Brazil and Europe. The project deploys the EU-Brazil Hybrid Data Infrastructure that allows the sharing of hardware, software and data on-demand. This infrastructure provides access to several integrated services and resources to seamlessly aggregate taxonomic, biodiversity and climate data, used by processing services implementing checklist cross-mapping and ecological niche modelling. A Virtual Research Environment was created to provide users with a single entry point to processing and data resources. This article describes the architecture, demonstration use cases and experimental results.

show abstract

Section: E Computing Servicesmentioning

confidence: 99%

Supporting biodiversity studies with the EUBrazilOpenBio Hybrid Data Infrastructure

Amaral

Badía

Blanquer

et al. 2014

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

show abstract

“…Efforts have been made at Universidade Federal Fluminense (UFF) , itero 'i, Brazil to make 1\/IPI system-aware to deal with its innate inefficiency towards heterogeneous systems . It resulted in the develop1nent of EasyGrid which is designed as an Application l\/Ianagen1ent System (AMS) for the MPI library [91].…”

Section: Easygridmentioning

confidence: 99%

“…Wit h t his init ial information , an initial scheduling schem a is prepared. At runt ime, t he dynamic sub-scheduling syste1n 1 nake runt i1ne cheduling decisions on t he basis of runt ime feedback [91].…”

Section: Easygridmentioning

confidence: 99%

Service Oriented Approach to High Performance Scientific Computing

Mulerikkal

Strazdins

2010

2010 10th IEEE/ACM International Conference on Cluster, Cloud and Grid Computing

View full text Add to dashboard Cite

Service Oriented Architectures (SOA) are used in business circles to harness the distributed resources of an enterprise, with relative ease and convenience. SOA is p erceived to deliver progran1n1ability, scalability and efficiency under heterogeneity. The scientific community is always aspired to h ave a similar convenient approach to solve "not so embarrassingly parallel" scientific problems with expected levels of high performance, especially in heterogeneous conditions. One of the major challenges in parallelizing scientific algorithms is their apparent interdep endency of tasks ( aton1ic units of parallel works) which results in too fine granularity. The computational advantages in parallelizing those scientific algorithms can b e overshadowed by costs involved in communications between non-opti1nal fine-grained tasks , in a SOA environment.The aim of this PhD research is to overcome these challenges and to empower scientists and researchers with SOA tools to develop high p erformance scientific applications •with relative ease that can perform well under heterogeneous envi-ronn1ents. The research has produced a scalable and heterogeneity-oblivious SOA middle,;vare -ANU-SOANI. It implements a popular enterprise SOA n1iddleware API (IB 1-Platform Syn1phony API) and thus ensures programmability. It offers better performance under heterogeneous conditions by imple1nenting load balancing and cheduling techniques. Along with its compute services it provides a Data Ser, ice ,;vhich helps application programmers to develop codes t hat can effectively circun1vent t he interdep endenc} of tasks and thereby reduce communications to ensure high perfonnance outcomes. The Data Service achieves t his by allo,;ving data to be stored accessed , n1odified and synchronized ( using add; get put and sync functionalities) at host and compute nodes according to the application logic. It is also obser, ed that the programming 1nodel supp orted by t he Data Service can help A -SOAl\lI applications to access compute resources in a Cloud IaaS over high latency networks (like t he Internet) with much lower overheads co1n pared to t he convent ional SOA program1ning models.

show abstract

“…Research at the Universidade Federal Fluminense (UFF), Nitero'i, Brazil in this line has resulted in the development of EasyGrid. It is designed as an Application Management System (AMS) for MPI library [15]. It allows applications to be autonomous, which is not a property of pure MPI applications with the help of a Resource Management System (RMS) [16] who implements self-scheduling policies.…”

Section: Related Workmentioning

confidence: 99%

“…It allows applications to be autonomous, which is not a property of pure MPI applications with the help of a Resource Management System (RMS) [16] who implements self-scheduling policies. EasyGrid has given positive results according to [17] and [15]. However, it is not clear that EasyGrid addresses the problems associated with task granularity and high communication costs expected in scientific applications.…”

Section: Related Workmentioning

confidence: 99%

An SOA approach to high performance scientific computing: Early experiences

Mulerikkal

Strazdins

2010

2010 International Conference on High Performance Computing

View full text Add to dashboard Cite

Abstract-Service Oriented Architecture (SOA) has been embraced in enterprise computing for several years. The scientific community always felt the need of an SOA infrastructure not only with the convenience of enterprise SOA but also with expected level of high performance capabilities. Our research has produced an SOA middleware (ANU-SOAM) which supports an already popular enterprise SOA middleware API (Platform Symphony API) with the desired level of performance for scientific computations such as a Conjugate Gradient Solver. We have extended the compute services of ANU-SOAM with a common data service (CDS) between client and the service instances. The aim is to improve performance of applications by reducing communications or communication cost between the client and the service instances with the help of CDS. This is achieved by enabling tasks to perform a deferred put operation to the common data their service instances, with the results of the put operation only being visible to the next generation of tasks. These updates can be synchronised (committed) at CDS at the direction of the client. This property enables applications on ANU-SOAM to overcome latency of poor networks (or 'cloud') between client and service instances. Experimental results on a small Gigabit ethernet cluster show that, for the Conjugate Gradient Solver, the ANU-SOAM version suffers no appreciable performance loss over MPI versions and the CDS enhances N-Body Solver performance, with good scalability in both cases.

show abstract

EasyGrid Enabling of Iterative Tightly-Coupled Parallel MPI Applications

Cited by 7 publications

References 11 publications

Supporting biodiversity studies with the EUBrazilOpenBio Hybrid Data Infrastructure

Supporting biodiversity studies with the EUBrazilOpenBio Hybrid Data Infrastructure

Service Oriented Approach to High Performance Scientific Computing

An SOA approach to high performance scientific computing: Early experiences

Contact Info

Product

Resources

About