Modern database applications operate on massive data and support a range of complex queries, especially OLAP queries which are time-consuming. To accelerate query processing, a variety of methods for automatically verifying query equivalence have been proposed to avoid redundant executions of equivalent queries, mainly in a semantic sense. However, we have observed some queries that are not semantically equivalent also return the same tuples under the specific data distribution, which cannot be detected by most current automated verification of query equivalence. To deal with this issue, this paper proposes weak equivalence for identifying queries that are not semantically equivalent but produce the same results under the read-mostly scenarios such as OLAP. Specifically, for posed queries, we extract their filter condition expressions, which are then transformed into symbolic representations, namely first-order logic formulae. In terms of their partial order, i.e. containment relationship, we introduce Query Lattice, a novel structure that is constructed as a lattice which is partitioned into equivalence classes that are convex to answer queries if we determine they belong to the classes. The equivalence class enables stored queries to respond to future unseen queries so that redundant generation of query plan and execution can be bypassed. Experimental evaluation of Query Lattice built on top of a prevailing open-source DBMS, PostgreSQL shows that the maximum improvement that Query Lattice can achieve is 44.95 % over the original PostgreSQL, when running on the datasets of both TPC-H and TPC-H Skew benchmarks.

Abstract In big data analysis, a large quantity of OLAP and aggregate queries exists, which have much stronger semantic context relationships (e.g. drill down and roll up) than generic SQL queries. Caching query results in memory playing an important role in accelerating data queries. Nevertheless, traditional query caching schemata neither fully utilize the features of OLAP, such as drill down and roll up semantics, nor compress the cached results, as the memory space is limited. In this paper, we propose a succinct, adaptive semantic OLAP caching, where the cache items are the cube lattice equivalence classes with only the bounds in a class stored. With further queries, the bound ranges are extended or expanded, indicating more query-answering ability which is assessed by the proposed covering capacity. The bounds of equivalence classes that more covering capacity are preferentially preserved in caching. We further empower our cache with some inference ability to derive more new data cells without posing extra queries and develop efficient query and update algorithms. The extensive experimental evaluation is conducted on synthetic and real data sets with various parameter settings. Our cache outperforms the common caching like LRU and LFU. Furthermore, it is robust to the non-repeated-pattern queries, still with a 30% hit ratio.