Kickfire: the SQL chip boosts the MySQL data warehouse to equip every MySQL database server with an SQL chip. this is a vision of Kickfire to focus on the data warehouse market.
Kickfire launched the Kickfire Database Appliance test version integrated with the new SQL chip in April 14. The company said that the SQL chip can directly extract data from the memory without the need to extract data from registers or cache, so it can alleviate the I/O bottleneck resulting in slow data query. This product is expected to be available in the market this fall.
With Kickfire technology, you can divide an SQL query into parallel query plans and send them to the SQL chip so that they can process data in parallel. When the queried data is returned from the memory in a compressed format, the data will flow into the SQL chip and be processed during the inbound process.
Steve Dille, vice president of marketing at Kickfire, said that the way SQL chips connect to existing hardware is similar to how graphics chips connect to the master server. The SQL chip is like a sub-processor on the master server. The SQL chip has a built-in concurrency that enables the application software to meet the load speed of multiple CPUs. In addition, Kickfire provides the incremental load function for MySQL to track changes in the source database and then automatically pass these changes to the Kickfire application.
Optimize query processing and data warehouse
Although the data warehouse market is dominated by vendors such as IBM, Oracle, and Teradata, open-source databases in Greenplum (PostgreSQL-based) and Infobright (working with MySQL) and other companies have developed their own territory. Wayne Eckerson, tdiffusion's research director, believes that Kickfire technology will help accelerate MySQL's application in the data warehouse market. MySQL has never optimized query processing and data warehouse before, so this is a huge step forward. In the past one and a half years, we have seen a lot of query accelerators entering the market and quietly retired, they try to improve the query performance of a data warehouse in a way that has been proven to be hasty. It takes some time to verify whether Kickfire can break this gloomy curse.
Dille said that Kickfire decided to focus on the MySQL market for technical and commercial reasons. Most commercial database applications start with open-source databases such as PostgreSQL, and constantly modify their code until it becomes essentially another private database completely different from the original database. MySQL supports plug-in storage engines in MySQL standards. Kickfire may be an integral part of the standard database. Therefore, we are still working hard on the SQL chip and database kernel in MySQL to maintain technological innovation. In terms of business, MySQL is the world's most popular open-source database, and most users have not been familiar with high-performance database applications such as Kickfire, so the market potential is still huge.
Henry Morris, an IDC analyst, believes that targeting the target market for customers who have adopted MySQL as a data warehouse solution gives Kickfire a great opportunity for development. The combination of integrated software and the most popular open source database will undoubtedly attract a lot of market attention