Book ChapterDOI
DyRecT: Software Support for Adaptive Parallelism on NOWs
Etienne Godard,Sanjeev Setia,Elizabeth L. White +2 more
- pp 1168-1175
TLDR
DyRecT (Dynamic Reconfiguration Toolkit) is a software library that allows programmers to develop adaptively parallel message-passing MPI programs for clusters of workstations and provides support for making a wider variety of applications adaptive by exposing to the programmer a low-level library that implements many of the typical tasks performed during reconfiguration.Abstract:
In this paper, we describe DyRecT (Dynamic Reconfiguration Toolkit) a software library that allows programmers to develop adaptively parallel message-passing MPI programs for clusters of workstations. DyRecT provides a high-level API that can be used for writing adaptive parallel HPF-like programs while hiding most of the details of the dynamic reconfiguration from the programmer. In addition, DyRecT provides support for making a wider variety of applications adaptive by exposing to the programmer a low-level library that implements many of the typical tasks performed during reconfiguration. We present experimental results for the overhead of dynamic reconfiguration of several benchmark applications using DyRecT.read more
Citations
More filters
Journal ArticleDOI
SRS: A framework for developing malleable and migratable parallel applications for distributed systems
Sathish Vadhiyar,Jack Dongarra +1 more
TL;DR: A framework for developing malleable and migratable MPI message-passing parallel applications for distributed systems that includes a user-level checkpointing library called SRS and a runtime support system that manages the checkpointed data for distribution to distributed locations is discussed.
Book ChapterDOI
Dynamic reconfiguration of grid-aware applications in ASSIST
Marco Aldinucci,Alessandro Petrocelli,Edoardo Pistoletti,Massimo Torquati,Marco Vanneschi,Luca Veraldi,Corrado Zoccolo +6 more
TL;DR: This work discusses a novel approach used in the ASSIST programming environment to implement/guarantee user provided QoS contracts in a transparent and effective way based on the implementation of automatic run-time reconfiguration of ASSIST application executions triggered by mismatch between the user providedQoS contract and the actual performance values achieved.
Journal ArticleDOI
Dynamicity in distributed applications: issues, problems and the ASSIST approach
Marco Vanneschi,L. Veraldi +1 more
TL;DR: The approach to dynamicity is described in the context of the ASSIST programming environment and it is shown how it is able to provide users with many of the required abstractions to develop adaptive, high-performance, distributed applications.
Journal ArticleDOI
Loosely coordinated coscheduling in the context of other approaches for dynamic job scheduling: a survey
TL;DR: This survey presents and compares the different approaches, while particularly focusing on the less well‐explored loosely coordinated time sharing, in terms of modification of standard operating systems, the runtime system and the communication libraries.
Book
Self-configuring and self-optimizing grid components in the GCM model and their ASSIST implementation
Marco Aldinucci,Carlo Bertolli,Sonia Campa,Massimo Coppola,Marco Vanneschi,Luca Veraldi,Corrado Zoccolo +6 more
TL;DR: The proposal of a super-component feature is part of the experience gained in the implementation of the ASSIST environment, which allows to develop self-configuring and optimising, component-based applications following a structured and hierarchical approach.
References
More filters
Journal ArticleDOI
Adaptive parallelism and Piranha
TL;DR: The authors show that adaptive parallelism has the potential to integrate heterogeneous platforms seamlessly into a unified computing resource and to permit more efficient sharing of traditional parallel processors than is possible with current systems.
Proceedings ArticleDOI
The utility of exploiting idle workstations for parallel computation
TL;DR: The system support required for adaptive parallel programs is developed and used to build an adaptive parallel computational fluid dynamics application and to answer the workstation-availability questions.
Journal ArticleDOI
Interfacing Condor and PVM to harness the cycles of workstation clusters
Jim Pruyne,Miron Livny +1 more
TL;DR: A framework for interfacing PVM, a popular system for parallel programming with Condor, a powerful resource management system is developed, and this combined system is operational, and further developments are made to provide a single coherent environment.
Proceedings ArticleDOI
Data parallel programming in an adaptive environment
TL;DR: This paper develops a runtime library to provide runtime support for data parallel programming in an adaptive environment and presents performance results for a multiblock Navier-Stokes solver run on a network of workstations using PVM for message passing.
Proceedings ArticleDOI
Transparent adaptive parallelism on NOWs using OpenMP
TL;DR: A system that allows OpenMP programs to execute on a network of workstations with a variable number of nodes that uses a DSM as the underlying substrate and insure that most adaptations can occur at these adaptation points, and thus at low cost.
Related Papers (5)
SRS: A framework for developing malleable and migratable parallel applications for distributed systems
Sathish Vadhiyar,Jack Dongarra +1 more