Browse Prior Art Database

Parallelize individual database queries to fully leverage hardware capabilities

IP.com Disclosure Number: IPCOM000236930D
Publication Date: 2014-May-22
Document File: 1 page(s) / 32K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method that allows a single query for data to be broken down into many concurrent queries to leverage the full capabilities of hardware.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 100% of the total text.

Page 01 of 1

Parallelize individual database queries to fully leverage hardware capabilities

The novel contribution is a method that allows a single query for data to be broken down into many concurrent queries to leverage the full capabilities of hardware.

The implementation steps follow:


1. Decompose a single time series search into N smaller time series searches


2. Execute all the time series searches in parallel


3. As the smaller queries finish, reconcile the results into a single result set

In an example embodiment, consider a database server with eight query processors. A single search arrives and the system decomposes it into eight distinct smaller queries. Those smaller queries can all run in parallel.

Now consider a second query arrives while the first query is being executed by the eight dedicated query processors. The database server reclaims half of the processors and assigns those to the second query. Now, both queries are running with four concurrent query processors.

If eight distinct queries all arrive for execution at the same time, then the database server allocates processors to queries on a 1:1 ratio per current processes. When any single query completes, the allocated processors are automatically assigned to any new or currently executing queries.

1