A Dollar from 15 Cents: Cross-Platform Management for Internet Services
Stewart, Christopher; Kelly, Terence; Zhang, Alex; Shen, Kai
Keyword(s): performance modeling, performance prediction, queueing models, CPU cache models processor cache models, multicore processors, cross-platform performance, prediction, processor selection, capacity planning, system management
Abstract: As Internet services become ubiquitous, the selection and management of diverse server platforms now affects the bottom line of almost every firm in every industry. Ideally, such cross-platform management would yield high performance at low cost, but in practice, the performance consequences of such decisions are often hard to predict. In this paper, we present an approach to guide cross-platform management for real-world Internet services. Our approach is driven by a novel performance model that predicts application-level performance across changes in platform parameters, such as processor cache sizes, processor speeds, etc., and can be calibrated with data commonly available in today's production environments. Our model is structured as a composition of several empirically observed, parsimonious sub- models. These sub-models have few free parameters and can be calibrated with lightweight passive observations on a current production platform. We demonstrate the usefulness of our cross-platform model in two management problems. First, our model provides accurate performance predictions when selecting the next generation of processors to enter a server farm. Second, our model can guide platform-aware load balancing across heterogeneous server farms.
Additional Publication Information: Published and presented at the USENIX Annual Technical Conference (USENIX'08). Boston, MA, June 2008.
External Posting Date: January 21, 2009 [Fulltext]. Approved for External Publication
Internal Posting Date: January 21, 2009 [Fulltext]