Faster Multidimensional Data Queries on Infrastructure Monitoring Systems

The analytics in online performance monitoring systems have often been limited due to the query of large scale multidimensional data. In this paper, we introduce a faster approach using bit-sliced index (BSI). Our study covers grouping and preference top-k queries with BSI, algorithms design, time complexity evaluation, comparison on real-time production system. research work extended BSI cover attributes filtering grouping. We evaluated single attribute, multiple attributes, feature filtering, To compare existing prior arts, made benchmarking bitmap indexing, sequential scan, collection streaming result our experiments data, proposed outperforms arts: 3 times than indexing attribute queries, 10 stream While comparing baseline scan approach, algorithm factor 100 queries. previous research, had space simulation data various distributions, further studied, evaluated, concluded real