UNIX network programming
Communications of the ACM
Limits to low-latency communication on high-speed networks
ACM Transactions on Computer Systems (TOCS)
PVM: Parallel virtual machine: a users' guide and tutorial for networked parallel computing
PVM: Parallel virtual machine: a users' guide and tutorial for networked parallel computing
ISCA '90 Proceedings of the 17th annual international symposium on Computer Architecture
Application-level scheduling on distributed heterogeneous networks
Supercomputing '96 Proceedings of the 1996 ACM/IEEE conference on Supercomputing
Multimethod communication for high-performance metacomputing applications
Supercomputing '96 Proceedings of the 1996 ACM/IEEE conference on Supercomputing
Design and implementation of multicast operations for ATM-based high performance computing
Proceedings of the 1994 ACM/IEEE conference on Supercomputing
CANPC '98 Proceedings of the Second International Workshop on Network-Based Parallel Computing: Communication, Architecture, and Applications
Exploiting multiple heterogeneous networks to reduce communication costs in parallel programs
HCW '97 Proceedings of the 6th Heterogeneous Computing Workshop (HCW '97)
Utilizing Heterogeneous Networks in Distributed Parallel Computing Systems
HPDC '97 Proceedings of the 6th IEEE International Symposium on High Performance Distributed Computing
A Performance Comparison of TCP/IP and MPI on FDDI, Fast Ethernet, and Ethernet
A Performance Comparison of TCP/IP and MPI on FDDI, Fast Ethernet, and Ethernet
Scatter and gather operations on an asynchronous communication model
SAC '00 Proceedings of the 2000 ACM symposium on Applied computing - Volume 2
Separated high-bandwidth and low-latency communication in the cluster interconnect Clint
Proceedings of the 2002 ACM/IEEE conference on Supercomputing
Journal of Parallel and Distributed Computing
Separated high-bandwidth and low-latency communication in the cluster interconnect clint
Separated high-bandwidth and low-latency communication in the cluster interconnect clint
Hi-index | 0.01 |
The different types of messages used by a parallel application program executing in a distributed computing system can each have unique characteristics so that no single communication network can produce the lowest latency for all messages. For instance, short control messages may be sent with the lowest overhead on one type of network, such as Ethernet, while bulk data transfers may be better suited to a different type of network, such as Fibre Channel or HiPPI. This work investigates how to exploit multiple heterogeneous communication networks that interconnect the same set of processing nodes using a set of techniques we call performance-based path determination (PBPD) [9], [10]. The performance-based path selection (PBPS) technique selects the best (lowest latency) network among several for each individual message to reduce the communication overhead of parallel programs. The performance-based path aggregation (PBPA) technique, on the other hand, aggregates multiple networks into a single virtual network to increase the available bandwidth. We test the PBPD techniques on a cluster of SGI multiprocessors interconnected with Ethernet, Fibre Channel, and HiPPI networks using a custom communication library built on top of the TCP/IP protocol layers. We find that PBPS can reduce communication overhead in applications compared to using either network alone, while aggregating networks into a single virtual network can reduce communication latency for bandwidth-limited applications. The performance of the PBPD techniques depends on the mix of message sizes in the application program and the relative overheads of the networks, as demonstrated in our analytical models.