Summary cache: a scalable wide-area Web cache sharing protocol
Proceedings of the ACM SIGCOMM '98 conference on Applications, technologies, architectures, and protocols for computer communication
Exploiting IP multicast in content-based publish-subscribe systems
IFIP/ACM International Conference on Distributed systems platforms
Practical network support for IP traceback
Proceedings of the conference on Applications, Technologies, Architectures, and Protocols for Computer Communication
Space/time trade-offs in hash coding with allowable errors
Communications of the ACM
WOSP '02 Proceedings of the 3rd international workshop on Software and performance
IEEE/ACM Transactions on Networking (TON)
JEWEL: Design and Implementation of a Distributed Measurement System
IEEE Transactions on Parallel and Distributed Systems
Exactly-once Delivery in a Content-based Publish-Subscribe System
DSN '02 Proceedings of the 2002 International Conference on Dependable Systems and Networks
Peer-to-peer overlay broker networks in an event-based middleware
Proceedings of the 2nd international workshop on Distributed event-based systems
A peer-to-peer approach to providing QoS monitoring for web service activities
Proceedings of the ACM/IFIP/USENIX Middleware '08 Conference Companion
Hi-index | 0.00 |
As enterprise applications rely increasingly on commodity messaging middleware, message tracking has become instrumental in testing and run-time monitoring. However, building an effective message tracking system is challenging because of the large scale and high message rate of enterprise-wide applications. To address this challenge, we consider the case of message tracking for distributed messaging middleware. We desire to record the origin, path, and destination of every application message while imposing low overhead with respect to latency, memory and storage. To achieve our goal, we propose a tunable approximation approach based on Bloom filter "histories." Our approach is tunable in the sense that more accurate audit trails may be provided at the expense of storage space, or, conversely, storage overhead is reduced for applications requiring less accurate audit trails. We describe the design of the system and demonstrate its utility by analyzing the performance of a prototype implementation.