Details
| Originalsprache | Englisch |
|---|---|
| Titel des Sammelwerks | CODES+ISSS '11: Proceedings of the seventh IEEE/ACM/IFIP international conference on Hardware/software codesign and system synthesis |
| Seiten | 119-128 |
| Seitenumfang | 10 |
| Publikationsstatus | Veröffentlicht - Okt. 2011 |
| Extern publiziert | Ja |
| Veranstaltung | Embedded Systems Week 2011, ESWEEK 2011 - 9th IEEE/ACM International Conference on Hardware/Software-Codesign and System Synthesis, CODES+ISSS'11 - Taipei, Taiwan Dauer: 9 Okt. 2011 → 14 Okt. 2011 |
Abstract
The trend towards many-core systems comes with various issues, among them their highly dynamic and non-predictable workloads. Hence, new paradigms for managing resources of many-core systems are of paramount importance. The problem of resource management, e.g. mapping applications to processor cores, is NP-hard though, requiring heuristics especially when performed online. In this paper, we therefore present a novel resource-management scheme that supports so-called malleable applications. These applications can adopt their level of parallelism to the assigned resources. By design, our (decentralized) scheme is scalable and it copes with the computational complexity by focusing on local decision-making. Our simulations show that the quality of the mapping decisions of our approach is able to stay near the mapping quality of stateof- the-art (i.e. centralized) online schemes for malleable applications but at a reduced overall communication overhead (only about 12,75% on a 1024 core system with a total workload of 32 multi-threaded applications). In addition, our approach is scalable as opposed to a centralized scheme and therefore it is practically useful for employment in large many-core systems as our extensive studies and experiments show.
ASJC Scopus Sachgebiete
- Informatik (insg.)
- Hardware und Architektur
- Informatik (insg.)
- Software
- Ingenieurwesen (insg.)
- Steuerungs- und Systemtechnik
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
CODES+ISSS '11: Proceedings of the seventh IEEE/ACM/IFIP international conference on Hardware/software codesign and system synthesis. 2011. S. 119-128.
Publikation: Beitrag in Buch/Bericht/Sammelwerk/Konferenzband › Aufsatz in Konferenzband › Forschung › Peer-Review
}
TY - GEN
T1 - DistRM: Distributed Resource Management for On-Chip Many-Core Systems
AU - Kobbe, Sebastian
AU - Bauer, Lars
AU - Lohmann, Daniel
AU - Schröder-Preikschat, Wolfgang
AU - Henkel, Jörg
PY - 2011/10
Y1 - 2011/10
N2 - The trend towards many-core systems comes with various issues, among them their highly dynamic and non-predictable workloads. Hence, new paradigms for managing resources of many-core systems are of paramount importance. The problem of resource management, e.g. mapping applications to processor cores, is NP-hard though, requiring heuristics especially when performed online. In this paper, we therefore present a novel resource-management scheme that supports so-called malleable applications. These applications can adopt their level of parallelism to the assigned resources. By design, our (decentralized) scheme is scalable and it copes with the computational complexity by focusing on local decision-making. Our simulations show that the quality of the mapping decisions of our approach is able to stay near the mapping quality of stateof- the-art (i.e. centralized) online schemes for malleable applications but at a reduced overall communication overhead (only about 12,75% on a 1024 core system with a total workload of 32 multi-threaded applications). In addition, our approach is scalable as opposed to a centralized scheme and therefore it is practically useful for employment in large many-core systems as our extensive studies and experiments show.
AB - The trend towards many-core systems comes with various issues, among them their highly dynamic and non-predictable workloads. Hence, new paradigms for managing resources of many-core systems are of paramount importance. The problem of resource management, e.g. mapping applications to processor cores, is NP-hard though, requiring heuristics especially when performed online. In this paper, we therefore present a novel resource-management scheme that supports so-called malleable applications. These applications can adopt their level of parallelism to the assigned resources. By design, our (decentralized) scheme is scalable and it copes with the computational complexity by focusing on local decision-making. Our simulations show that the quality of the mapping decisions of our approach is able to stay near the mapping quality of stateof- the-art (i.e. centralized) online schemes for malleable applications but at a reduced overall communication overhead (only about 12,75% on a 1024 core system with a total workload of 32 multi-threaded applications). In addition, our approach is scalable as opposed to a centralized scheme and therefore it is practically useful for employment in large many-core systems as our extensive studies and experiments show.
KW - Manycore
KW - Mpsoc
KW - Multi-agent-system
KW - Multicore
KW - Resource management
KW - Scalability
UR - http://www.scopus.com/inward/record.url?scp=81355142727&partnerID=8YFLogxK
U2 - 10.1145/2039370.2039392
DO - 10.1145/2039370.2039392
M3 - Conference contribution
AN - SCOPUS:81355142727
SN - 9781450307154
SP - 119
EP - 128
BT - CODES+ISSS '11: Proceedings of the seventh IEEE/ACM/IFIP international conference on Hardware/software codesign and system synthesis
T2 - Embedded Systems Week 2011, ESWEEK 2011 - 9th IEEE/ACM International Conference on Hardware/Software-Codesign and System Synthesis, CODES+ISSS'11
Y2 - 9 October 2011 through 14 October 2011
ER -