Abstract Accurate network monitoring is vital for the operation of Grids. The packet loss ratio is among the most important metrics for identifying poor network conditions, since it highly affects data throughput performance and the overall end-to-end data transfer quality. In this paper, we present a scalable and non-intrusive technique based on passive network monitoring for estimating the packet loss ratio between different measurement points. The proposed approach is complementary to current active monitoring techniques and can be easily incorporated into the network monitoring components of Grid systems. We describe the design and implementation of the technique, outline its integration within a Grid environment, and present experimental evaluation results, including measurements with real Grid application traffic.

We report on the design, implementation, and evaluation of a system called Cedar that enables mobile database access with good performance over low-bandwidth networks. This is accomplished without degrading consistency. Cedar exploits the disk storage and processing power of a mobile client to compensate for weak connectivity. Its central organizing principle is that even a stale client replica can be used to reduce data transmission volume from a database server. The reduction is achieved by using content addressable storage to discover and elide commonality between client and server results. This organizing principle allows Cedar to use an optimistic approach to solving the difficult problem of database replica control. For laptop-class clients, our experiments show that Cedar improves the throughput of read-write workloads by 39 % to as much as 224 % while reducing response time by 28 % to as much as 79%.

opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF or Carnegie Mellon University. All unidentified trademarks mentioned in the paper are properties of their respective owners.

Dimorphic computing is a new model of computing that switches between thick and thin client modes of execution in a completely automated and transparent manner. It accomplishes this without imposing any language or structural requirements on applications. This model greatly improves the performance of applications that alternate between phases of compute- or data-intensive processing and intense user interaction. For such applications, the thin client mode allows efficient use of remote resources such as compute servers or large datasets. The thick client mode enables crisp interactive performance by eliminating the harmful e#ects of Internet latency and jitter, and by exploiting local graphical hardware acceleration.

their respective owners.

unidentified trademarks mentioned in the paper are properties of their respective owners. Keywords: Content Addressable Techniques, Content Addressable Storage, Network