Timo: In-Memory Temporal Query Processing for Big Temporal Data

Abstract

We present Timo, a distributed in-memory temporal query and analytics model, to process mass intensive temporal data efficiently. Firstly, Timo implements a set of efficient execution processes for big temporal data. Secondly, a novel temporal index is suggested that has more efficiently query performance and less memory storage than the state-of-art methods. Thirdly, according to the temporal locality feature of temporal query, Timo suggests a partitioner mechanism, named FSpartitioner, which improves query throughput. Lastly, we deploy Timo system on Apache Spark platform, and experimental results show that Timo is superior to other Spark based temporal systems in respect of both query latency and throughput.