This paper considers the question of how effective caches are in processing network I/O. Our analysis shows that operating system structure plays a key role in the caches behavior, with BSD Unix (a monolithic OS) making more effective use of the cache than Mach (a microkernel OS). Moreover, closer inspection shows that several factors contribute to this result, including how TLBs are managed, h...

In the reordering buffer problem (RBP), a server is asked to process a sequence of requests lying in a metric space. To process a request the server must move to the corresponding point in the metric. The requests can be processed slightly out of order; in particular, the server has a buffer of capacity k which can store up to k requests as it reads in the sequence. The goal is to reorder the r...

We study photonic packet switching architecture that enables a high node through-put and provides priority services in DiffServ (Differentiated Services). As we will show in this paper, a photonic packet switch can be equipped with an ultra-high-speed address-lookup capability in the optical domain. However, the optical buffer must be managed in the electrical domain. In this paper, we investig...

1209 wire choices are available, because only the look-up table needs to be reconstructed for different choices of wire sizes. The look-up table is constructed very efficiently since every buffer-to-buffer delay is computed incrementally and only once. In other words, as the number of wire sizes increases, the new method is able to gain reduction in delay without need of dramatically increasing...

Supervisory control for deadlock-free resource allocation has been an active area of manufacturing systems research. Most work, however, assumes that allocated resources do not fail. Little research has addressed allocating resources that may fail. Automated manufacturing systems have many types of components that may fail unexpectedly. We develop robust controllers for single unit resource all...

In this paper we propose storage allocation schemes to minimize the buffer requirements for fault-tolerant video-ondemand (VOD) servers with disk arrays. To guarantee the high reliability of video service, the servers usually adopt parity-encoding techniques in which data blocks and their associated parity block form a parity group. For real-time video service, all the blocks in a parity group ...

Reliable multicast delivery requires that a multicast message be received by all members in a group. Hence certain or all members need to buffer messages for possible retransmissions. Designing an efficient buffer management algorithm is challenging in large multicast groups where no member has complete group membership information and the delivery latency to different members could differ by o...

FERRIC SULFATE BY HUMIDITY BUFFER TECHNIQUE AND RAMAN SPECTROSCOPY AT 0.1 MPa. W. G. Kong, Alian Wang and I-Ming Chou Dept. Earth and Planetary Sciences and McDonnell Center for Space Sciences, Washington University, St. Louis, MO 63130, USA ([email protected]), 2 School of Space Science and Physics, Shandong University, Weihai, 264200, China. 3 954 National Center, U.S. Geological Survey,...