Abstract

When users submit new queries to a distributed stream processing system (DSPS), a query planner must allocate physical resources from a set of hosts to queries. In this talk I will discuss about query planning in two different settings.

First, I will talk about query planning that targets DSP Ss where allocation decisions must provide the correct mix of resources required by queries for their processing. I will describe SQPR , a query planner which exploits overlap between queries and reuses partial results to save resources and increase the number of admitted queries. SQPR models query admission, allocation, and reuse as a single constrained optimisation problem and solves an approximate version to achieve scalability.

Second, I will refer to a DSPS system that focuses on fair query processing in resource starved deployments. Internet-scale stream processing systems for processing data from millions of sensors will have to support ever-increasing numbers of queries and data sources. Given a finite amount of resources, such systems are likely to suffer from constant overload, thus only processing partial data. I will describe FISSP , a query planning framework which accepts new queries in overload conditions, by transferring resources from existing queries to new ones through controlled and fair degradation of processing quality.

Bio: Eva Kalyvianaki is a Post-Doctoral Researcher at the Department of Computing, Imperial College London. Before this, she obtained her PhD from the Computer Lab, University of Cambridge. Her main research interests are in systems, distributed systems management, and autonomic computing. http://www.doc.ic.ac.uk/~ekalyv