| LHCb Note | |
| Report number | LHCb-2005-081 ; CERN-LHCb-2005-081 |
| Title | Etude d'architectes de méta-ordonnancement pour le calcul en régime permanent et saturé |
| Translation of title | A study of meta-scheduling architectures for high throughput computing |
| Author(s) | Caron, E ; Garonne, V ; Tsaregorodtsev, A Yu |
| Affiliation | (LIP/ENS, Lyon) ; (CPPM, Marseille) |
| Series | (Offline) |
| Submitted to | 4th International Symposium on Parallel and Distributed Computing, Lille, France, 4 - 6 Jul 2005 |
| Submitted by | Sibylle.Greiner@cern.ch on 02 Feb 2006 |
| Subject category | Detectors and Experimental Techniques |
| Accelerator/Facility, Experiment | CERN LHC ; LHCb |
| Abstract | In this paper we present a model and simulator for many clusters of heterogeneous PCs belonging to a local network. These clusters are assumed to be connected to each other through a global network and each cluster is managed via a local scheduler which is shared by many users. We validate our simulator by comparing the experimental and analytical results of a M/M/4 queuing system. These studies indicate that the simulator is consistent. Next, we do the comparison with a real batch system and we obtain an average error of 10.5% for the response time and 12% for the makespan. We conclude that the simulator is realistic and well describes the behaviour of a large-scale system. Thus we can study the scheduling of our system called dirac in a high throughput context. We justify our decentralized, adaptive and opportunistic approach in comparison to a centralized approach in such a context. |
| Copyright/License | Preprint: (License: CC-BY-4.0) |
Corresponding record in: Inspire
Journalen skapades 2006-02-02, och modifierades senast 2018-06-11
