SQL query method and system for Presto

Abstract

The invention provides an SQL query method and system for Presto. The method comprises: firstly, configuring a table in which a to-be-cached query result is located in a database; secondly, according to an SQL query command, generating an executive plan and generating a unique identifier for the executive plan; and finally, according to the unique identifier of the executive plan, querying a cache table cached in a distributed file system and obtaining a query result in the cache table. According to the SQL query method and system, when the same SQL query is executed in a period of time, Presto calculation is no longer executed and a previously cached query result is directly returned, so that the query speed is increased, the Presto pressure is reduced, and the consumption of a memory and a CPU is reduced.

Description

technical field [0001] The invention belongs to the field of interactive SQL query, in particular to an SQL query method and system for Presto. Background technique [0002] Presto is a pure memory interactive SQL query engine. The query efficiency is nearly 10 times higher than that of Hive (a data warehouse tool based on Hadoop, which provides simple SQL query functions). Multiple data sources are used to query, and it has been successfully applied to well-known Internet companies such as Facebook, Netflix, Dropbox, and JD.com. [0003] For the SQL submitted to Presto for execution, Presto first parses the SQL and generates an execution plan, and then assigns and executes computing tasks according to the execution plan, such as figure 1 shown. [0004] During the use of Presto, sometimes the following problems are encountered: [0005] 1. SQL will be executed frequently; [0006] 2. The execution time is long but the query result is small; [0007] 3. The real-time pe...

AI Technical Summary

Problems solved by technology

[0005] 1. SQL will be executed frequently;

[0006] 2. The execution time is long but the query result is small;

[0007] 3. The real-time performance of query results is not high

However, due to the long execution time of each SQL, the pressure on Presto calculation is greater, and the consumption of memory and CPU is also greater, which may even lead to memory overflow and SQL execution failure.

Images