Getting ready for approximate computing: Trading parallelism for accuracy for dss workloads

Abstract

Processors have evolved dramatically in the last years and current multicore systems deliver very high performance. We are observing a rapid increase in the number of cores per processor thus resulting in more dense and powerful systems. Nevertheless, this evolution will meet several challenges such as power consumption, and reliability. It is expected that, in order to improve the effciency, future processors will contain units that are able to operate at a very low power consump-tion with the drawback of not guaranteeing the correctness of the produced results. This model is known as Approxi-mate Computing. One interesting approach to exploit Ap-proximate Computing is to make applications aware of the errors and react accordingly. For this work we focus on the Decision Support System Workloads and in particular the standard TPC-H set of queries. We rst dene a metric that quanties the cor-rectness of a query result -Quality of Result (QoR). Using this metric we analyse the impact of relaxing the correct-ness in the DBMS on the accuracy of the query results. In order to improve the accuracy of the results we propose a dynamic adaptive technique that is implemented as a tool above the DBMS. Using heuristics, this tool spawns a num-ber of replica query executions on different cores and com-bines the results as to improve the accuracy. We evalu-ated our technique using real TPC-H queries and data on PostgreSQL with a simple fault-injection to emulate the Ap-proximate Computing model. The results show that for the selected scenarios, the proposed technique is able to increase the QoR with a cost in parallel resources smaller than any alternative static approach. The results are very encourag-ing since the QoR is within 7% of the best possible. © is held by the owner/authors.