Browse Prior Art Database

A system and method to build an SQL Interface for cloud databases with efficient query optimizer

IP.com Disclosure Number: IPCOM000205545D
Publication Date: 2011-Mar-30
Document File: 6 page(s) / 245K

Publishing Venue

The IP.com Prior Art Database

Related People

V. Bharat: INVENTOR [+4]

Abstract

The present disclosure provides an SQL system that uses the storage and scaling capabilities of cloud. The method provided supports importing current relational data on to the cloud. The system provided has a join cost based optimizer which can execute the SQL queries more efficiently by dynamic selection of early and late materialization of tuples. The system provided, accesses Column store databases of Cloud, transform the SQL query over row-store data into an equivalent query over column-store data which is of type key-value storage in order to retrieve the data from cloud databases. This disclosure proposes using one common implementation interface which is a modified version of Structured Query Language (SQL) to interact in a platform independent manner, thus make it more generic in nature. It also builds efficient cost based optimizer that reduces the number of selection and join operations over column-store databases.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 34% of the total text.

Page 01 of 6

A system and method to build an SQL Interface for cloud databases with efficient query optimizer Field of Disclosure

This disclosure relates to database management for cloud computing.

Background of the disclosure

Cloud databases predominantly store key-value pairs, as they are based on the Column-store databases.

Since most of the data is now being migrated to cloud, different custom commands to upload and

download data to and from various clouds is discouraging users. Applications need to be completely re- written for using data over cloud. This task is not only time consuming and costly, but it also introduces

new bugs and troubles as the developer may not be well equipped to use write API's for each of the

cloud the user needs to access.

The present technology mostly addresses traditional relational databases which follow row-store.

However, cloud databases which are relational nature follow column-store. Few organizations like

Amazon provide APIs to interact with the cloud data. Most of these APIs are specific to their own cloud

databases, but not provide a generic APIs to interact with them. For example, Hadoop (from Apache)

provides its own syntax which differs from the syntax from Amazon EC2. On the other hand, they mostly

focus on inserting data, rather than providing an SQL kind of interface to retrieve the data. Further, there is no open standard for cloud data interfaces. So, if the cloud infrastructure is using

differing libraries to interact, code needs to be re-written to each of the library separately. So, hosting

an existing enterprise-based application over cloud has to take place again a complete software

development life cycle. Similar challenges arise during data migration from one cloud to the other. We

need to know about the interfaces first and then develop specific migration tools which are very time

consuming process.

Abstract

The present disclosure provides an SQL system that uses the storage and scaling capabilities of cloud.

The method provided supports importing current relational data on to the cloud. The system provided

has a join cost based optimizer which can execute the SQL queries more efficiently by dynamic selection

of early and late materialization of tuples. The system provided, accesses Column store databases of

Cloud, transform the SQL query over row-store data into an equivalent query over column-store data

which is of type key-value storage in order to retrieve the data from cloud databases. This disclosure

proposes using one common implementation interface which is a modified version of Structured Query

Language (SQL) to interact in a platform independent manner, thus make it more generic in nature. It

also builds efficient cost based optimizer that reduces the number of selection and join operations over

column-store databases.


Page 02 of 6

Drawings

Fig 1 shows the flowchart for the process.

Detailed Description

Figure 1 shows the flowchart of present invention. Brief description of v...