Elasticity Manager for Elastic Key-Value Stores in the Cloud
Abstract:
The increasing spread of elastic Cloud services, together with the pay-as-you-go pricing model of Cloud computing, has led to the need of an elasticity controller. The controller automatically resizes an elastic service in response to changes in workload, in order to meet Service Level Objectives (SLOs) at a reduced cost. However, variable performance of Cloud Virtual Machines and nonlinearities in Cloud services, such as the diminishing reward of adding a service instance with increasing the scale, complicates the controller design. First, we briefly discuss challenges and some approaches to automation of elasticity of a cloud-based storage, and, then, present the design and evaluation of ElastMan, an elasticity controller for Cloud-based elastic key-value stores. ElastMan combines feedforward and feedback control. Feedforward control is used to respond to spikes in the workload by quickly resizing the service to meet SLOs at a minimal cost. Feedback control is used to correct modeling errors and to handle diurnal workload. To address nonlinearities, our design of ElastMan leverages the near-linear scalability of elastic Cloud services in order to build a scale-independent model of the service. We have implemented and evaluated ElastMan using the Voldemort key-value store running in an OpenStack Cloud environment.
Our evaluation shows the feasibility and effectiveness of our approach to automation of Cloud service elasticity.
Biography:
Vladimir Vlassov is an associate professor of computer systems at the Department of Software and Computer Systems, School of Information and Communication Technology, KTH Royal Institute of Technology in Stockholm, Sweden. Before joining KTH in 1993, he was an assistant professor and an associate professor at the Department of Computer Engineering, Electro-technical University of St. Petersburg, Russia (1985-1993), from which he obtained a doctoral degree in computer science (1984). In 1998, he worked as a visiting scientist at Laboratory for Computer Science, Massachusetts Institute of Technology (MIT), Cambridge, MA, USA. In 2004 he worked as an invited visiting researcher at Department of Electrical and Computer Engineering, University of Massachusetts (UMASS), Amherst, MA, USA. He has experience in national and international research projects and a long teaching experience. He is currently participating in a number of research projects in the area of Cloud computing and multi-core systems. At KTH, he earned a docent degree in computer systems in 2006. His current research focus is on data-intensive computing and stream processing; resource management in multi-clouds, cloud federations and inter-cloud; self-management of cloud-based services and applications (including auto-scaling, optimization of data placement and data-intensive computation); system and workload modeling.