MemSQL, a new breed of databases has gone live today that uses an abstract framework that combines processing speed and robust SQL interfce within an in-memory data tier to achieve actionable insights at rates 30 times faster than with disk-backed alternatives.
MemSQL places data into memory and translates SQL into C++ for the utmost optimization in query execution. This enables MemSQL to write and read data at incredible speeds, and by offering a relational interface, you can unify the data you’d normally store in a short-lived medium—cache or key-value store—and place it directly into a database along with your existing data.
This in-memory data store achieves the speed it claims with a unique concept where a query when fired from the system comes across a linear scan parser that strips of any numbering or string parameters and saves a skeleton of the query in plan cache. Any future query with the same query skeleton is accelerated directly through a hot code path to the pre-compiled execution engine.
MemSQL leverages a combination of lock-free data structures with an innovative coding translation process to efficiently retrieve and process data in memory. The solution is ideal for applications that require fast processing of machine data, including financial services, digital advertising, bioinformatics, government, manufacturing, and mobile applications.
MemSQL also announced today that the company has secured a total of $5 million in funding to date. The company’s investors are a combination of prominent venture capitalists and angels, including First Round Capital, IA Ventures, NEA, SV Angel, Y Combinator, Paul Buchheit, Ashton Kutcher, Max Levchin and Aaron Levie. MemSQL intends to use the funds to scale the company’s infrastructure across sales, marketing and technical support to take advantage of market opportunities where fast analysis of machine data is crucial.