HOMA: automatic re-scheduling of multiple invocations in CORBA

Thierry Gautier 1 Hamid-Reza Hamidi 1
1 APACHE - Parallel algorithms and load sharing
ID-IMAG - Informatique et Distribution, Inria Grenoble - Rhône-Alpes, UJF - Université Joseph Fourier - Grenoble 1
Abstract : This report presents HOMA, an efficient and scalable CORBA-based code coupling environment. HOMA is composed by an IDL compiler and a runtime support. From IDL definitions of CORBA objects, HOMA compiler automatically extracts all the informations which allow efficient and scalable composition of method invocations. The compiler and runtime support rely on two functionalities: the automatic extraction of parallelism between method invocations and the lazy protocol used to communicate effective parameters. The runtime re-schedules the invocations and the associated communications using the parallelism deduced from their IDL definitions. The scheduling strategy is based on ATHAPASCAN. Used together these functionalities enable parallel communication between code coupled applications. This property is conserved by composition of invocations. The technology is based on the generation of extended client stub and server skeleton on top of standard client stub and server skeleton generated by existing IDL/CORBA compiler. Parallelism is handled by the parallel programming environment ATHAPASCAN. Thus, HOMA is highly portable. The target applications are high performance numerical simulations.
Document type :
Reports
Complete list of metadatas

https://hal.inria.fr/inria-00070801
Contributor : Rapport de Recherche Inria <>
Submitted on : Friday, May 19, 2006 - 9:40:34 PM
Last modification on : Wednesday, April 11, 2018 - 1:55:38 AM
Long-term archiving on : Sunday, April 4, 2010 - 8:17:31 PM

Identifiers

  • HAL Id : inria-00070801, version 1

Collections

Citation

Thierry Gautier, Hamid-Reza Hamidi. HOMA: automatic re-scheduling of multiple invocations in CORBA. [Research Report] RR-5191, INRIA. 2004, pp.24. ⟨inria-00070801⟩

Share

Metrics

Record views

245

Files downloads

721