Assessing the Performance of Communication Layer in a Cluster of Workstations
Abstract:
Abstract---Recent improvements in LANs make network of workstations a good alternative to traditional parallel computers in some applications. However, in this platform the communication performance is over two orders of magnitude inferior to state-of-art multiprocessors. Currently networking technologies have put the pressure in the software overhead. Because of this, applications could not take advantage of this communication performance potential. In this paper, we present an implementation of Virtual Circuit Caching that reduces the software overhead by allocating communication resource once and re-using them for multiple messages to the same destinations. With this approach, the communication overhead is reduced by approximately a 35 % for long messages; this reduction should enable the extensive use of networks of workstations for scientific parallel applications.
Citations
| 232 | BEOWULF: A parallel workstation for scienti£c computation – Sterling, Savarese, et al. - 1995 |
| 158 | A.: Using MPI – Gropp, Lusk, et al. - 1994 |
| 49 | Assessing Fast Network Interfaces – Culler, Liu, et al. - 1996 |
| 26 | Eicken et al., Active Messages: a Mechanism for Integrated Communication and Computation – von - 1992 |
| 12 | Architectural support for reducing communication overhead in multiprocessor interconnection networks – Dao, Yalamanchili, et al. - 1997 |
| 9 | Analyzing the performance of MPI in a cluster of workstations based on fast Ethernet – PIERNAS, FLORES, et al. - 1997 |
| 2 | Optimal Communicationn Performance on Fast Ethernet with GAMMA – Ciaccio - 1998 |
| 1 | A low Latency Communication System for Fast Ethernet – Verma, Chiueh, et al. - 1998 |

