Coasters: Uniform Resource Provisioning and Access for Clouds and Grids

Hategan, Mihael; Wozniak, Justin M.; Maheshwari, Ketan

doi:10.1109/ucc.2011.25

Cited by 41 publications

(35 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Coasters framework in Hategan et al (2011), offers both usability and performance goals; in term of usability, by build a uniform framework that allows a multiple services system to access into cloud resources. In other word, it is a uniform access and resources provisioning for cloud and grid.…”

Section: Dynamic Resource Provisioning and Monitoring In Cloud Computingmentioning

confidence: 99%

“…In other word, it is a uniform access and resources provisioning for cloud and grid. The jobs in Coasters framework runs independent (dependences between jobs), that allow to apply a parallel computing paradigm for the jobs by using the Swift language (Hategan et al, 2011) a scripting language oriented to scientific computing which can automatically parallelize the execution of scripts and distribute tasks to various resources. So, the users do not require to log into remote systems and prepare services or configure compute nodes.…”

Section: Dynamic Resource Provisioning and Monitoring In Cloud Computingmentioning

confidence: 99%

See 1 more Smart Citation

Towards improving resource management in cloud systems using a multi-agent framework

Al‐Ayyoub

Daraghmeh

Jararweh

et al. 2016

IJCC

View full text Add to dashboard Cite

Abstract:With the goal of efficient sharing of resources and services, the cloud computing paradigm has gained a lot of interest recently. This work focuses on improving the resource utilisation by optimising the resource provisioning through a multi-agent framework in which different agents are responsible for different tasks including the monitoring of customers (behaviour, resource usage patterns and QoS requirements as stated in the SLA) and available resources based on customer's requests. Moreover, we introduce the concept of TaskFlow which allows a more elastic resources provisioning to match the customer real usage of the resources. The proposed system is implemented and tested on the CloudSim simulator and the results show it increases resource utilisation and decreases power consumption while avoiding SLA violations. The results also show that the introduction of the concept of TaskFlow into our proposed system leads to more resource saving but with a higher risk of SLA violations.

show abstract

Section: Dynamic Resource Provisioning and Monitoring In Cloud Computingmentioning

confidence: 99%

Section: Dynamic Resource Provisioning and Monitoring In Cloud Computingmentioning

confidence: 99%

Towards improving resource management in cloud systems using a multi-agent framework

Al‐Ayyoub

Daraghmeh

Jararweh

et al. 2016

IJCC

View full text Add to dashboard Cite

show abstract

“…In this mode, a Swift interpreter, written in Java, executes on a single (possibly multicore) host, and sends work to distributed systems using a variety of "providers" that interface with remote systems. When using its own resource provisioner [6] Swift is capable of sending approximate 500 tasks per second to a set of resources. When using Falkon [3], Swift achieved over 1,000 tasks per second.…”

Section: Swift: Dataflow Execution and Programming Model For Mtcmentioning

confidence: 99%

“…In recent years we have extended MTC to applications on homogeneous supercomputers, using tools such as Falkon [3], Swift [4], JETS [5], and Coasters [6]. Other programming models and tools that support MTC include MapReduce, volunteer computing [7], SLURM [8], and Cobalt [9], which allow supercomputer tasks to be subdivided into asynchronous subtasks [10].…”

Section: Introductionmentioning

confidence: 99%

Design and evaluation of the gemtc framework for GPU-enabled many-task computing

Krieder

Wozniak

Armstrong

et al. 2014

Proceedings of the 23rd International Symposium on High-Performance Parallel and Distributed Computing

View full text Add to dashboard Cite

We present the design and first performance and usability evaluation of GeMTC, a novel execution model and runtime system that enables accelerators to be programmed with many concurrent and independent tasks of potentially short or variable duration. With GeMTC, a broad class of such "many-task" applications can leverage the increasing number of accelerated and hybrid high-end computing systems. GeMTC overcomes the obstacles to using GPUs in a many-task manner by scheduling and launching independent tasks on hardware designed for SIMD-style vector processing. We demonstrate the use of a high-level MTC programming model (the Swift parallel dataflow language) to run tasks on many accelerators and thus provide a highproductivity programming model for the growing number of supercomputers that are accelerator-enabled. While still in an experimental stage, GeMTC can already support tasks of fine (subsecond) granularity and execute concurrent heterogeneous tasks on 86,000 independent GPU warps spanning 2.7M GPU threads on the Blue Waters supercomputer.

show abstract

“…Swift [1] is an application level scripting framework designed for composing ordinary programs into parallel applications. Applications encoded in U.S. Government work not protected by U.S. copyright Swift have been shown to execute on multiple computational sites via Swift coasters [2], [3] mechanism that implements the pilot jobs paradigm. Swift provides a simple reactive resource scheduling wherein, based on an initial "wave" of jobs, it records the per site job completion rate and adjusts the propor tionate number of jobs to be sent to these sites.…”

Section: Introductionmentioning

confidence: 99%

Model-driven multisite workflow scheduling

Maheshwari

Jung

Meng

et al. 2013

2013 IEEE International Conference on Cluster Computing (CLUSTER)

View full text Add to dashboard Cite

Abstract-Workflows continue to play an important role in expressing and deploying scientific applications. In recent years, a wide variety of computational sites have emerged with shared access to users. A user may not be able to complete a complex workflow at a single site. It is thus beneficial to run different tasks of a workflow on different sites. For such cases, judicious scheduling strategy is required in order to map tasks in the workflow to resources at multiple sites so that the workload is balanced among sites and the overhead is minimized in data transfer. The key challenge is that the data transfer rate among sites varies based on the network capacity and load.We propose a workflow scheduling technique that tackles the multi-site task distribution challenge by using data movement performance modeling. We applied this technique to schedule an earth observation science workflow over three sites. Executed via the Swift parallel scripting paradigm, we augmented its default schedule and improved the time-to-completion by up to 52%.

show abstract

Coasters: Uniform Resource Provisioning and Access for Clouds and Grids

Cited by 41 publications

References 19 publications

Towards improving resource management in cloud systems using a multi-agent framework

Towards improving resource management in cloud systems using a multi-agent framework

Design and evaluation of the gemtc framework for GPU-enabled many-task computing

Model-driven multisite workflow scheduling

Contact Info

Product

Resources

About