An Automated Profiling Subsystem for QoS-Aware Services

The advent of QoS-sensitive Internet applications such as multimedia, the inception of priced online services, and the proliferation of performance-critical applications such as online trading raise a need for building server systems with guaranteed performance. Unlike embedded software systems designed for a particular special-purpose platform, the new applications should be designed to run on diierent heterogeneous platforms, provide soft performance guarantees commensurate with platform capacity, and adapt eeciently to upgrades in platform resources over the system's lifetime. The problem of prooling the application for the purposes of providing QoS guarantees on each new platform becomes a signiicant undertaking. Automated prooling mechanisms must be built to enable eecient computing of QoS guarantees tailored to platform capacity and facilitate wide deployment of soft performance-guaranteed systems on heterogeneous platforms. In this paper, we investigate the design of the automated prooling subsystem; an essential component of future \general-purpose" QoS-sensitive systems. The subsystem estimates application resource requirements and adapts the software transparently to the resource capacity of the underlying platform. A novel aspect of the proposed prooling subsystem is its use of estimation theory for prooling. Resource requirements are estimated by correlating applied workload with online resource utilization measurements. We focus explicitly on prooling server software. The convergence and accuracy of our online prooling techniques are evaluated in the context of an Apache web server serving both static web pages and dynamic content. Results show the viability of using estimation theory for automated online prooling and for achieving QoS guarantees.