Loading...
Please wait, while we are loading the content...
Similar Documents
Abstract user-level communication in cluster-based servers ∗.
| Content Provider | CiteSeerX |
|---|---|
| Author | Bianchini, Ricardo Rao, Srinath Carrera, Enrique V. Iftode, Liviu |
| Abstract | Clusters of commodity computers are currently being used to provide the scalability required by several popular Internet services. In this paper we evaluate an efficient cluster-based WWW server, as a function of the characteristicsof the intra-cluster communication architecture. More specifically, we evaluate the impact of processor overhead,network bandwidth, remote memory writes, and zero-copy data transfers on the performance of our server. Our experimental results with an 8-node cluster and four real WWW traces show that network bandwidth affects the performanceof our serverby only 6%. In contrast, user-level communication can improve performance by as much as 29%. Low processor overhead, remote memory writes, and zero-copyall make small contributions towardsthis overallgain. To be able to extrapolate from our experimental results, we use an analytical model to assess the performance of our server under different workload characteristics, different numbers of cluster nodes, and higher performance systems. Our modeling results show that higher gains (of up to 55%) can be accrued for workloads with large working sets and next-generation servers running on large clusters. 1 |
| File Format | |
| Access Restriction | Open |
| Subject Keyword | Real Www Trace Zero-copy Data Transfer Low Processor Overhead Next-generation Server Efficient Cluster-based Www Server Analytical Model Different Workload Characteristic Processor Overhead Abstract User-level Communication User-level Communication 8-node Cluster Zero-copyall Make Small Contribution Large Cluster Performance System Commodity Computer Remote Memory Writes Cluster-based Server Network Bandwidth Intra-cluster Communication Architecture Experimental Result Several Popular Internet Service Different Number Cluster Node |
| Content Type | Text |
| Resource Type | Article |