For service applications on a network, measuring availability, performance, and quality of service is critical. Yet traditional software and hardware measures are both inadequate and misleading. Better measures of availability that incorporate end-user experience will lead to meaningful benchmarks and progress in providing high availability services. We present the results of a series of long-term experiments that measured availability of select Web sites and services with the goal of duplicating the end-user experience. Using our measurements, we propose a new metric for availability that goes beyond the traditional sole measure of uptime.

Clusters of PCs are increasingly popular as cost-effective platforms for supercomputer-class applications. Given recent performance improvements in graphics accelerators, clusters are similarly attractive for demanding graphics applications. We describe the design and implementation of Lightning-2, a display subsystem for such a cluster. The system scales in both the number of rendering nodes and the number of displays supported, and allows any pixel data generated from any node to be dynamically mapped to any location on any display. A number of image-compositing functions are supported, including color-keying and depth-compositing. A distinguishing feature of the system is its platform independence: it connects to graphics accelerators via an industry-standard digital video port and requires no modifications to accelerator hardware or device drivers. As a result, rendering clusters that utilize Lightning-2 can be upgraded across multiple generations of graphics accelerators with little effort. We demonstrate a renderer that achieves 106 Mtri/s on an 8-node cluster using Lightning-2 to perform sort-last depth compositing.