Search for: All records

The Standard Performance Evaluation Corporation (SPEC) CPU benchmark has been widely used as a measure of computing performance for decades. The SPEC is an industry-standardized, CPU-intensive benchmark suite and the collective data provide a proxy for the history of worldwide CPU and system performance. Past efforts have not provided or enabled answers to questions such as, how has the SPEC benchmark suite evolved empirically over time and what micro-architecture artifacts have had the most influence on performance? - have any micro-benchmarks within the suite had undue influence on the results and comparisons among the codes? - can the answers to these questions provide insights to the future of computer system performance? To answer these questions, we detail our historical and statistical analysis of specific hardware artifacts (clock frequencies, core counts, etc.) on the performance of the SPEC benchmarks since 1995. We discuss in detail several methods to normalize across benchmark evolutions. We perform both isolated and collective sensitivity analyses for various hardware artifacts and we identify one benchmark (libquantum) that had somewhat undue influence on performance outcomes. We also present the use of SPEC data to predict future performance. 

Designing modern web applications involves a wide spectrum of choices when it comes to deciding where the different tiers of application and framework code that constitute these distributed applications should be placed. These system design choices affect programmer productivity, ease of deployment, security, and performance, particularly with respect to latency and scalability. In this paper, we propose and evaluate a design choice in which not only all application logic executes server-side, but most presentation logic as well. The client browser is reduced to a rendering and I/O engine, similar to a "thin client" or \dumb terminal," but retains the full expressiveness of rich, modern Internet applications. We have developed CloudBrowser 2.0, a system that implements this distribution model using a scalable multiprocess approach. In this paper, we perform an evaluation of the benefits and costs of this approach when compared to both more traditional approaches as well as emerging alternatives. We focus on programmability and systems aspects including performance and latency.