Abstract
In the multiprocessors adopting ring topology with high-speed unidirectional point-to-point links, the performance of microprocessors, local busses and memory systems continue to improve at a very fast rate, whereas bandwidth of point-to-point links does not increase. Therefore ring network configured by current off-the-shelf point-to-point links cannot meet bandwidth requirements for system level interconnection, which may eventually limit scalability of system. For this reason, necessity to expand bandwidth of ring interconnect is emphasized. In this paper, we focus on the bandwidth of the interconnection network in a ring-based CC-NUMA system with snooping protocol. We show that ring interconnect is a bottleneck of performance in the high-end environment, if each node(cluster) is connected by point-to-point links based on the IEEE Scalable Coherent Interface(SCI) specifications. Bandwidth expansion by more than 200% is needed to achieve scalable performance. In order to expand ring bandwidth, it does not seem a cost effective way to develop a new link with doubled data line width. As an alternative to doubling data line width, we propose a dual ring architecture. In this paper, several ways to implement dual ring are suggested: simple dual ring, transaction-separated dual ring and direction-separated dual ring. We analyze pros. and cons. of each model compared with wide single ring (which means single ring using links with doubled data line width) model and we also simulate each model with On-Line Transaction Processing (OLTP) workloads. This paper demonstrates that dual ring architecture, in spite of some defects against wide single ring, scales system performance comparable to wide single ring, and that direction-separated dual ring outperforms wide single ring when there are large number of processors in the system.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the 2001 ACM Symposium on Applied Computing, SAC 2001 |
| Publisher | Association for Computing Machinery |
| Pages | 510-516 |
| Number of pages | 7 |
| Volume | Part F129805 |
| ISBN (Print) | 1581132875, 9781581132878 |
| DOIs | |
| Publication status | Published - 2001 Mar 1 |
| Externally published | Yes |
| Event | 2001 ACM Symposium on Applied Computing, SAC 2001 - Las Vegas, United States Duration: 2001 Mar 11 → 2001 Mar 14 |
Other
| Other | 2001 ACM Symposium on Applied Computing, SAC 2001 |
|---|---|
| Country | United States |
| City | Las Vegas |
| Period | 01/3/11 → 01/3/14 |
Fingerprint
Keywords
- Interconnection networks
- Parallel architecture
- Ring bandwidth
- Ring-based CC-NUMA
ASJC Scopus subject areas
- Software
Cite this
Efficient schemes to scale the interconnection network bandwidth in a ring-based multiprocessor system. / Jang, Byoung Soon; Jung, Sung Woo; Jhang, Seong Tae; Jhon, Chu Shik.
Proceedings of the 2001 ACM Symposium on Applied Computing, SAC 2001. Vol. Part F129805 Association for Computing Machinery, 2001. p. 510-516.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Efficient schemes to scale the interconnection network bandwidth in a ring-based multiprocessor system
AU - Jang, Byoung Soon
AU - Jung, Sung Woo
AU - Jhang, Seong Tae
AU - Jhon, Chu Shik
PY - 2001/3/1
Y1 - 2001/3/1
N2 - In the multiprocessors adopting ring topology with high-speed unidirectional point-to-point links, the performance of microprocessors, local busses and memory systems continue to improve at a very fast rate, whereas bandwidth of point-to-point links does not increase. Therefore ring network configured by current off-the-shelf point-to-point links cannot meet bandwidth requirements for system level interconnection, which may eventually limit scalability of system. For this reason, necessity to expand bandwidth of ring interconnect is emphasized. In this paper, we focus on the bandwidth of the interconnection network in a ring-based CC-NUMA system with snooping protocol. We show that ring interconnect is a bottleneck of performance in the high-end environment, if each node(cluster) is connected by point-to-point links based on the IEEE Scalable Coherent Interface(SCI) specifications. Bandwidth expansion by more than 200% is needed to achieve scalable performance. In order to expand ring bandwidth, it does not seem a cost effective way to develop a new link with doubled data line width. As an alternative to doubling data line width, we propose a dual ring architecture. In this paper, several ways to implement dual ring are suggested: simple dual ring, transaction-separated dual ring and direction-separated dual ring. We analyze pros. and cons. of each model compared with wide single ring (which means single ring using links with doubled data line width) model and we also simulate each model with On-Line Transaction Processing (OLTP) workloads. This paper demonstrates that dual ring architecture, in spite of some defects against wide single ring, scales system performance comparable to wide single ring, and that direction-separated dual ring outperforms wide single ring when there are large number of processors in the system.
AB - In the multiprocessors adopting ring topology with high-speed unidirectional point-to-point links, the performance of microprocessors, local busses and memory systems continue to improve at a very fast rate, whereas bandwidth of point-to-point links does not increase. Therefore ring network configured by current off-the-shelf point-to-point links cannot meet bandwidth requirements for system level interconnection, which may eventually limit scalability of system. For this reason, necessity to expand bandwidth of ring interconnect is emphasized. In this paper, we focus on the bandwidth of the interconnection network in a ring-based CC-NUMA system with snooping protocol. We show that ring interconnect is a bottleneck of performance in the high-end environment, if each node(cluster) is connected by point-to-point links based on the IEEE Scalable Coherent Interface(SCI) specifications. Bandwidth expansion by more than 200% is needed to achieve scalable performance. In order to expand ring bandwidth, it does not seem a cost effective way to develop a new link with doubled data line width. As an alternative to doubling data line width, we propose a dual ring architecture. In this paper, several ways to implement dual ring are suggested: simple dual ring, transaction-separated dual ring and direction-separated dual ring. We analyze pros. and cons. of each model compared with wide single ring (which means single ring using links with doubled data line width) model and we also simulate each model with On-Line Transaction Processing (OLTP) workloads. This paper demonstrates that dual ring architecture, in spite of some defects against wide single ring, scales system performance comparable to wide single ring, and that direction-separated dual ring outperforms wide single ring when there are large number of processors in the system.
KW - Interconnection networks
KW - Parallel architecture
KW - Ring bandwidth
KW - Ring-based CC-NUMA
UR - http://www.scopus.com/inward/record.url?scp=84948801231&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84948801231&partnerID=8YFLogxK
U2 - 10.1145/372202.372447
DO - 10.1145/372202.372447
M3 - Conference contribution
AN - SCOPUS:84948801231
SN - 1581132875
SN - 9781581132878
VL - Part F129805
SP - 510
EP - 516
BT - Proceedings of the 2001 ACM Symposium on Applied Computing, SAC 2001
PB - Association for Computing Machinery
ER -