Today: May 22, 2018, 2:26 am
  
Computer & Technology

When SQLite hits the wall, multithreaded DBMS are ready to resolve the performance botteneck.

PR-Inside.com: 2018-05-04 16:58:40
SQLite is one of the most popular database management systems out there and with good reason. However, we can’t overlook the fact that there is a significant loss in performance when there are multiple users of the database writing at the same time.

With over a trillion SQLite databases deployed, millions of application owners are forced to look for alternative embedded databases like Raima Database Manager to resolve their performance bottleneck.


In this test we explain why users run into the performance bottlenecks and alternative options to overcome them.

The test is based on a standard TPC-B test framework that measures throughput in terms of how many transactions per second the system can perform.

Find the test here and learn how to overcome low performance in database management.

The majority of today’s operating system and hardware support multithreading. SQLite does not take advantage of this opportunity because of the design in SQLite. When there are many concurrent writes to a SQLite database, application users experience a significant reduction in speed and the application may not meet the users performance expectations. It is well known that the write access to the SQLite database can only be granted if no other requests are being serviced. It is the "one write at a time" design within SQLite that slows down the throughput. Thus, many application owners are forced to look for alternative embedded database options to resolve their performance bottleneck.

Raima has put together a test that proves the differences between the RDM database solution and SQLite. In this test, we demonstrate that RDM is a good alternative to SQLite.

How we did the test

In this performance comparison, we used a standard TPC-B  test framework from www.tpc.org. The TPC-B measures throughput in terms of how many transactions per second the system can perform. The test has been modified to allow for comparisons where multiple clients are doing parallel work.

We used the same test environment and framework for both SQLite and RDM. In the test, RDM proved to be significantly faster than SQLite.  See illustration below. The platform used for the test was RDM release 14.1 running against SQLite v3.23 on a standard Windows 10, Intel i7 processor machine with 16GB's of RAM and standard SATA hard drive.

RDM has a design that is optimized for embedded systems with a modern and efficient set of API's, along with a well-designed database storage file format. RDM allows for concurrent writes and reads and therefore can do many more times the amount of work that SQLite can do.

Press Information
Raima





email
www.raima.com

Published by
Fredrik Sande
2067485207
e-mail
raima.com



# 428 Words
Related Articles
 
More From Computer & Technology
Greeting dear players and investors!
After countless hours of hard work and determination we finally present the fully functional assets that we have all been [..]
The largest conference Blockchain Indo in Jakarta
Jakarta is one of the top 10 crypto-capitals of the world in 2017, according to IDACB (International Decentralised Association of [..]
Role of Cloud Service Providers In Cloud [..]
By Sachin Bhardwaj, Director Marketing & Business Development, eHosting DataFort Migration of data to the cloud is the process of [..]
UnicornGo – a new way of profitable [..]
Creators of the game paid particular attention to realistic genetics. In collaboration with genetic engineers, developers created [..]
Nexans wins 'Cabling Vendor of the Year’ [..]
Nexans, the worldwide leader in cables and cabling systems has been named the ‘Cabling Vendor of the Year 2018,’ at [..]

Disclaimer: If you have any questions regarding information in this press release please contact the company added in the press release. Please do not contact pr-inside. We will not be able to assist you. PR-inside disclaims the content included in this release.