Increased response time during periods of overload on a Web server may cause impatient users to time-out, causing the server to do unproductive work in processing these abandoned requests. Overhead time spent in preprocessing each request adds to the unproductive work even for requests that are not taken up for service. This causes the usable throughput, i.e. goodput, of the overloaded Web server to drop drastically, while resource utilization remains at 100%. Although this behaviour can be easily reproduced experimentally, existing analytical models of queues with abandonments are not able to do so.We present an analytical model that captures characteristics specific to networked software servers, namely, overhead processing and contention for shared hardware resources, that is able to explain the goodput degradation typically observed in overloaded servers. We use this model to compare the performance of the LIFO and FIFO queueing disciplines during overload and show that LIFO goodput andresponse time are better than those of FIFO.
展开▼
