DeepSQL: The Next Generation of Database Optimization

By Andy C.

Due to the explosive growth of Big Data in this world, SQL, the programming language for interacting and modifying databases, was also required to evolve in order to keep up. With such large amounts of data, servers needed to be expanded in order to accommodate all of the data. However, more servers would also mean more time to process data. A platform that can hold and process all of the data efficiently was needed. With databases being increasingly larger and increasingly more necessary for operations within a company, many companies needed to look for alternative solutions to handle the database for higher efficiency. Many different platforms to store and process data with different versions of SQL were developed, and a recent development in 2015 was the invention of DeepSQL by Deep Information Sciences. DeepSQL is a storage engine that uses MySQL, which transforms a MySQL database into a database that continually tunes itself for maximum optimization.

There are three steps to the DeepSQL engine. First, the CASSI (Continuously Adapting Sequential Summarization of Information) scans the hardware to determine the cost of usage while running. It also analyzes the data to check the volume and variety of data. After the CASSI figures out how the database currently works, it predicts and configures the database parameters to run the hardware, software, and algorithms of reading and writing data. After configuring the database, the CASSI continues to monitor and optimize the database as it is reading and writing data. CASSI continues to run this loop of analysis, adaption, and optimization, in order to run the database at a high efficiency. Not only is the database high in efficiency, but the database is also capable of handling great amounts of data.

Previously, databases would require a human to manually analyze and tune a database, and they would have to do that every time there were a sizable amount of changes in the database. They would also have to take into account the scalability of their settings, and configure the database to be able to handle more than it is currently handling. Humans can also make mistakes. Now, with DeepSQL, there is no need for someone to spend time configuring the database, and the database continuously changes the parameters on its’ own to maximize efficiency. Since DeepSQL uses MySQL, companies that use MySQL do not need to learn a new programming language, but they can use their existing MySQL database to run with the Deep Engine, which saves time on learning something new.

Even though DeepSQL was released less than a year ago, it is already gaining traction in the field. Amazon Web Services started offering DeepSQL to their customers as an option for running the cloud-based databases. Jones from Deep Information Sciences, in an interview, even claims that DeepSQL can outperform Amazon’s own customized version of MySQL, Aurora, by seven times.

DeepSQL also won multiple awards in 2015. One such award is the Gold award for “Best New Product of the Year – Enterprise” by Best in Biz. Best in Biz is an independent business awards program, and the fifth annual Best in Biz program had over 600 entries, judged by 46 judges. DeepSQL topped all of entries, and showed tremendous potential in the management of data.

There are companies that have started using DeepSQL in their system, and seeing outstanding results. One such company is Ruder Boskovic Institute (RBI), who are using the free Community Edition of DeepSQL for their genomic research. RBI has a massive amount of data with their research, and it takes a long time for their system to process and query all of that data. DeepSQL accelerates critical research processes by 400%, and is also ACID compliant, so nothing is lost if a server shuts down. DeepSQL enables RBI to ingest 50GB of genomes in four hours, which is 600% faster than InnoDB, and 200% faster than TokuDB. Not only does DeepSQL speed up the processes, but it also reduces the footprint of the database by 40% without impacting performance. Since DeepSQL is completely automated, they do not need to dedicate any researchers on loading and preprocessing data, which gives them more time to do the actual research.

Another company that started using DeepSQL is GEMServers. GEMServers is a high performance WordPress managed hosting business, and they needed a fast database in order to perform better than their competitors. By using DeepSQL, they are able to accelerate page loading by 200% and process transactions 39 times faster than using InnoDB. Since DeepSQL also cuts database size, it cuts down on bandwidth, which allows for them to save on bandwidth costs. DeepSQL is automated, which frees up employees to focus on the hosting operations instead of manual tuning of the database.

In all of the instances where DeepSQL was tested or implemented, the results were blowing every other solution away in performance. DeepSQL can change the landscape of the management of data through the data infrastructure, and, since its’ release in April of 2015, it will continue to grow. Deep Information Sciences are even offering DeepSQL Developer Edition (for testing and educational use) and Community Edition (for businesses with less than 1 million dollars in revenue) for free, which will allow for many people to use this technology and see its power. As DeepSQL becomes more prominent, this will prompt other solutions of the management of data to rise to compete, which will accelerate the development of the field of Big Data.

 

References

[Image of the process of the DeepSQL Engine]. (n.d.). Retrieved January 26, 2016, from https://deepis.com/How-Deep-Works

Alsultanny, Y. (n.d.). Database management and partitioning to improve database processing performance. Journal of Database Marketing & Customer Strategy Management, 17.3-4, 271-276. Retrieved January 28, 2016.

Babcock, C. (2015, September 29). DeepSQL: Containerized Database With New Performance Levels. Retrieved January 26, 2016, from http://www.informationweek.com/cloud/software-as-a-service/deepsql-containerized-database-with-new-performance-levels–/d/d-id/1322362

DeepIS. (n.d.). Retrieved January 26, 2016, from https://deepis.com/

DeepSQL earns gold award for “best new product of the year – enterprise” from best in biz. (2015, Dec 08). Business Wire Retrieved from http://search.proquest.com/docview/1746283555?accountid=10357

GEMServers selects DeepSQL to deliver hosted WordPress sites at breakthrough speed on google cloud platform. (2015, Nov 04). Business Wire Retrieved from http://search.proquest.com/docview/1729330712?accountid=10357

Noyes, K. (2015, November 10). How big data is changing the database landscape for good. Retrieved January 26, 2016, from http://www.pcworld.com/article/3003858/how-big-data-is-changing-the-database-landscape-for-good.html

Ruder boskovic institute selects DeepSQL community edition to accelerate genomic research at big data scale. (2015, Nov 19). Business Wire Retrieved from http://search.proquest.com/docview/1734273433?accountid=10357