Application execution management on the InteGrade opportunistic grid middleware

Silva, Francisco José da Silva e; Kon, Fábio; Goldman, Alfredo; Finger, Marcelo; Camargo, Raphael Y. de; Filho, Fernando Castor; Costa, Fábio M.

doi:10.1016/j.jpdc.2010.01.010

Cited by 11 publications

(4 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hence, there has been research for many years on Grid middleware solutions, e.g., OurGrid [42], InteGrade [120], to enable the execution of computationally-intensive applications on sets of geographically distributed clusters of machines. Originally targeting the scientific community with their large data processing needs, Grids can provide solutions to any computingintensive application.…”

Section: Research Challengesmentioning

confidence: 99%

Service-oriented middleware for the Future Internet: state of the art and research directions

Issarny

Georgantas

Hachem

et al. 2011

J Internet Serv Appl

143

View full text Add to dashboard Cite

Service-oriented computing is now acknowledged as a central paradigm for Internet computing, supported by tremendous research and technology development over the last 10 years. However, the evolution of the Internet, and in particular, the latest Future Internet vision, challenges the paradigm. Indeed, service-oriented computing has to face the ultra large scale and heterogeneity of the Future Internet, which are orders of magnitude higher than those of today's service-oriented systems. This article aims at contributing to this objective by identifying the key research directions to be followed in light of the latest state of the art. This article more specifically focuses on research challenges for service-oriented middleware design, therefore, investigating service description, discovery, access, and composition in the Future Internet of services.

show abstract

Section: Research Challengesmentioning

confidence: 99%

Service-oriented middleware for the Future Internet: state of the art and research directions

Issarny

Georgantas

Hachem

et al. 2011

J Internet Serv Appl

143

View full text Add to dashboard Cite

show abstract

“…mapper's address is stored in VMR's database (6). Each time a map result is validated, the VMR server checks if all map tasks have been executed and validated.…”

Section: Vmrmentioning

confidence: 99%

“…Desktop Grids may include enterprise environments, schools or scientific laboratories. This means that some Desktop Grids support more complex applications, such as MPI [5,6]. While a VC platform may be deployed on top of a desktop cluster, this is not the target environment intended for existing http://www.jisajournal.com/content/4/1 /18 systems.…”

Section: Introductionmentioning

confidence: 99%

Internet-scale support for map-reduce processing

Costa

Veiga

Ferreira

2013

J Internet Serv Appl

View full text Add to dashboard Cite

Volunteer Computing systems (VC) harness computing resources of machines from around the world to perform distributed independent tasks. Existing infrastructures follow a master/worker model, with a centralized architecture. This limits the scalability of the solution due to its dependence on the server. Our goal is to create a fault-tolerant VC platform that supports complex applications, by using a distributed model which improves performance and reduces the burden on the server. In this paper we present VMR, a VC system able to run MapReduce applications on top of volunteer resources, spread throughout the Internet. VMR leverages users' bandwidth through the use of inter-client communication, and uses a lightweight task validation mechanism. We describe VMR's architecture and evaluate its performance by executing several MapReduce applications on a wide area testbed. Our results show that VMR successfully runs MapReduce tasks over the Internet. When compared to an unmodified VC system, VMR obtains a performance increase of over 60% in application turnaround time, while reducing server bandwidth use by two orders of magnitude and showing no discernible overhead.

show abstract

“…Load value is calculated taking such information: CPU, memory resource, number of current tasks, response time and number of network links. Silva et al (da Silva e Silva et al, 2010) explained the resource management on the InteGrade grid middleware. They presented a grid as a collection of clusters, where each one runs its own Cluster Managers (CM).…”

Section: Related Workmentioning

confidence: 99%