Algorithmic skeletons meeting grids
Parallel Computing - Algorithmic skeletons
Securing skeletal systems with limited performance penalty: The muskel experience
Journal of Systems Architecture: the EUROMICRO Journal
Co-design of Distributed Systems Using Skeleton and Autonomic Management Abstractions
Euro-Par 2008 Workshops - Parallel Processing
Managing Multi-concern Application Complexity in AspectSBASCO
ICCS '09 Proceedings of the 9th International Conference on Computational Science: Part I
Grid'BnB: a parallel branch and bound framework for grids
HiPC'07 Proceedings of the 14th international conference on High performance computing
Exceptions for algorithmic skeletons
Euro-Par'10 Proceedings of the 16th international Euro-Par conference on Parallel processing: Part II
A survey of algorithmic skeleton frameworks: high-level structured parallel programming enablers
Software—Practice & Experience - Focus on Selected PhD Literature Reviews in the Practical Aspects of Software Technology
Joint structured/unstructured parallelism exploitation in muskel
ICCS'06 Proceedings of the 6th international conference on Computational Science - Volume Part II
Process-driven biometric identification by means of autonomic grid components
International Journal of Autonomous and Adaptive Communications Systems
Fine tuning algorithmic skeletons
Euro-Par'07 Proceedings of the 13th international Euro-Par conference on Parallel Processing
Management in distributed systems: a semi-formal approach
Euro-Par'07 Proceedings of the 13th international Euro-Par conference on Parallel Processing
| Hi-index | 0.00 |
We present an evolution of the Lithium parallel programming environment (the muskel Java package) that implements new features suitable for handling problems arising in typical dynamic environments such as the grid ones. In particular, our environment provides a prototype application manager that takes completely care of maintaining a given quality of service. The first version of muskel we discuss here is able to ensure a constant number of active processing elements assigned to computations in presence of failures of different types. The application manager embedded in muskel represents a prototype experience. It demonstrates the feasibility of a run time support controlled quality of service approach. In particular, it shows that a user supplied "performance contract" can be automatically ensured.