HTCaaS: Leveraging Distributed Supercomputing Infrastructures for Large-Scale Scientific Computing
Gespeichert in:
| Titel: | HTCaaS: Leveraging Distributed Supercomputing Infrastructures for Large-Scale Scientific Computing |
|---|---|
| Autoren: | Jik-soo Kim, Seungwoo Rho, Seoyoung Kim, Sangwan Kim, Seokkyoo Kim, Soonwook Hwang |
| Weitere Verfasser: | The Pennsylvania State University CiteSeerX Archives |
| Quelle: | http://datasys.cs.iit.edu/events/MTAGS13/p02.pdf. |
| Bestand: | CiteSeerX |
| Schlagwörter: | Many-Task Computing, HT- CaaS, Multi-level Scheduling, Dynamic Fairnes |
| Beschreibung: | In this paper, we present the HTCaaS (High-Throughput Computing as a Service) which aims to provide researchers with ease of exploring large-scale and complex scientific prob-lems by leveraging national supercomputing infrastructures in Korea. HTCaaS allows users to efficiently submit a large number of jobs at once by effectively managing and exploit-ing of all available computing resources. HTCaaS exploits a synthesis of well known techniques and its own intelligent scheduling algorithm to effectively support multiple users independently submitting large numbers of tasks to a col-lection of geographically distributed computing resources. Throughout our micro-benchmark and protein docking ex-periments, we show that our HTCaaS can provide a single efficient job management system that can support the most challenging scientific applications. |
| Publikationsart: | text |
| Dateibeschreibung: | application/pdf |
| Sprache: | English |
| Relation: | http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.646.9632; http://datasys.cs.iit.edu/events/MTAGS13/p02.pdf |
| Verfügbarkeit: | http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.646.9632 http://datasys.cs.iit.edu/events/MTAGS13/p02.pdf |
| Rights: | Metadata may be used without restrictions as long as the oai identifier remains attached to it. |
| Dokumentencode: | edsbas.F79B27B8 |
| Datenbank: | BASE |
Nájsť tento článok vo Web of Science | Abstract: | In this paper, we present the HTCaaS (High-Throughput Computing as a Service) which aims to provide researchers with ease of exploring large-scale and complex scientific prob-lems by leveraging national supercomputing infrastructures in Korea. HTCaaS allows users to efficiently submit a large number of jobs at once by effectively managing and exploit-ing of all available computing resources. HTCaaS exploits a synthesis of well known techniques and its own intelligent scheduling algorithm to effectively support multiple users independently submitting large numbers of tasks to a col-lection of geographically distributed computing resources. Throughout our micro-benchmark and protein docking ex-periments, we show that our HTCaaS can provide a single efficient job management system that can support the most challenging scientific applications. |
|---|