Burstiness in Multi-tiered Systems: Symptoms, Causes, and New Models
College of William and Mary - USA
DEI - Seminar Room
October 11th, 2010
Workload flows in enterprise systems that use the multi-tier paradigm are often characterized as bursty, i.e., exhibit a form of temporal dependence. Burstiness often results in dramatic degradation of the perceived user performance, which is extremely difficult to capture with existing capacity planning models. The main reason behind this deficiency of traditional capacity planning models is that the user perceived performance is the result of the complex interaction of a very complex workload with a very complex system.
In this talk, I will present a simple and effective methodology for detecting burstiness symptoms in multi-tier systems rather than identifying the low-level exact cause of burstiness as traditional models would require. This information on burstiness is incorporated into a surprisingly simple and effective new modeling methodology.
Detailed experimentation on a TPC-W testbed where all measurements are obtained by HP (Mercury) Diagnostics, a commercially available tool, shows that experimental and model prediction results are in excellent agreement and argue strongly for the effectiveness of the proposed methodology under both bursty and non-bursty workloads.
Beyond new models that capture burstiness, I will also present a new methodology for generating workloads that emulate the temporal surge/burstiness phenomenon in a controllable way, thus provide a mechanism that enables testing and evaluation of client-server system performance under reproducible bursty workloads. This methodology can be seamlessly incorporated into existing benchmarks that are routinely used for capacity planning of autonomic systems.
Area di ricerca: