Potential benefits of delta encoding and data compression for HTTP
SIGCOMM '97 Proceedings of the ACM SIGCOMM '97 conference on Applications, technologies, architectures, and protocols for computer communication
Internet Web servers: workload characterization and performance implications
IEEE/ACM Transactions on Networking (TON)
BLT: Bi-layer tracing of HTTP and TCP&slash;IP
Proceedings of the 9th international World Wide Web conference on Computer networks : the international journal of computer and telecommunications netowrking
Characteristics of WWW Client-based Traces
Characteristics of WWW Client-based Traces
The measured access characteristics of world-wide-web client proxy caches
USITS'97 Proceedings of the USENIX Symposium on Internet Technologies and Systems on USENIX Symposium on Internet Technologies and Systems
System design issues for internet middleware services: deductions from a large client trace
USITS'97 Proceedings of the USENIX Symposium on Internet Technologies and Systems on USENIX Symposium on Internet Technologies and Systems
Linux-HA heartbeat system design
ALS'00 Proceedings of the 4th annual Linux Showcase & Conference - Volume 4
Cost and Response Time Simulation forWeb-based Applications on Mobile Channels
QSIC '05 Proceedings of the Fifth International Conference on Quality Software
Performance tuning and cost discovery of mobile web-based applications
International Journal of Web Engineering and Technology
A Distributed Architecture for IP Traffic Analysis
AIMS '07 Proceedings of the 1st international conference on Autonomous Infrastructure, Management and Security: Inter-Domain Management
| Hi-index | 0.00 |
Web traffic has been increasing and evolving rapidly in recent years. It is important to measure the volume and characteristic of such dominant traffic to understand large-scale user access pattern and analyze performance of Web applications. Among the common methods of Web measurements, the passive way using packet monitoring is more advantageous since it provides comprehensive information and is transparent to end-users. However, the throughput of current packet monitoring system is limited by the bandwidth of network adapters. Computational capacity and buffer size are also potential performance bottlenecks for monitoring high-speed links. This paper proposes a cluster-based online monitoring system, which generates rich logs by packet sniffing for Web analysis in a cluster environment. Powered by cluster computing technologies, the system achieves high performance and availability. Other techniques, such as online reconstruction in memory and kernel packet filter, are also adopted to improve the system performance. The experimental results indicate that it is feasible to trace in high-speed links with the cluster-based system.