Engineering an autonomic partitioning framework for Grid-based SAMR applications
2004
Dynamic structured adaptive mesh refinement (SAMR) methods for the numerical solution to partial differential equations yield highly advantageous ratios for cost/accuracy when compared to methods based upon static uniform approximations. However, distributed Grid-based SAMR implementations present significant challenges. This chapter presents ARMaDA, an autonomic partitioning framework that provides adaptive, system and application sensitive partitioning, load balancing and configuration support to address these challenges. The overall goal of ARMaDA is to manage the dynamism and space-time heterogeneity of SAMR applications and Grid environments, and support the efficient and scalable execution of Grid-based SAMR implementations.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
26
References
4
Citations
NaN
KQI