Distributing Hot-Spot Addressing in Large-Scale Multiprocessors
IEEE Transactions on Computers
Performance of the Direct Binary n-Cube Network for Multiprocessors
IEEE Transactions on Computers
Interconnection networks for large-scale parallel processing: theory and case studies (2nd ed.)
Interconnection networks for large-scale parallel processing: theory and case studies (2nd ed.)
Algorithms for scalable synchronization on shared-memory multiprocessors
ACM Transactions on Computer Systems (TOCS)
Hierarchical hypercube multicomputer interconnection networks
Hierarchical hypercube multicomputer interconnection networks
Hot-Spot Contention in Binary Hypercube Networks
IEEE Transactions on Computers
Highly parallel computing (2nd ed.)
Highly parallel computing (2nd ed.)
On the effectiveness of combining in resolving “hot spot” contention
Journal of Parallel and Distributed Computing
Using a Multipath Network for Reducing the Effects of Hot Spots
IEEE Transactions on Parallel and Distributed Systems
Prevention of Congestion in Packet-Switched Multistage Interconnection Networks
IEEE Transactions on Parallel and Distributed Systems
The Use of Feedback in Multiprocessors and Its Application to Tree Saturation Control
IEEE Transactions on Parallel and Distributed Systems
Alleviating Memory Contention in Matrix Computations on Large-Scale Shared-Memory Multiprocessors
Alleviating Memory Contention in Matrix Computations on Large-Scale Shared-Memory Multiprocessors
Overview of Recent Supercomputers
Overview of Recent Supercomputers
DataStager: scalable data staging services for petascale applications
Proceedings of the 18th ACM international symposium on High performance distributed computing
DataStager: scalable data staging services for petascale applications
Cluster Computing
Dynamic evolution of congestion trees: analysis and impact on switch architecture
HiPEAC'05 Proceedings of the First international conference on High Performance Embedded Architectures and Compilers
Cost-efficient buffer sizing in shared-memory 3D-MPSoCs using wide I/O interfaces
Proceedings of the 49th Annual Design Automation Conference
Hi-index | 0.00 |
In parallel systems, it is possible for several processors to request concurrent access to a shared data structure such as a synchronization variable. Such an access pattern causes what is known as hot-spot contention. In shared-memory multiprocessor systems that use a multistage interconnection network, hot-spot contention can result in "tree saturation" that degrades the system performance. It is important, therefore, to manage hot-spot connection properly. This article reviews the existing strategies to reduce the effects of hot-spot contention. The authors first quantify the effects of hot-spot connection and identify the objectives of a hot-spot management strategy. They propose a taxonomy to categorize these strategies, and then they review and compare several representative strategies that have been proposed to reduce the effects of hot-spot contention.