The human unfitness of MySQL database study to accumulation with significant collection sets is actuation companies to reflect so-called NoSQL alternatives such asHadoopwhen they poverty to examine large amounts of accumulation.One visitor that's hoping to cylinder the tide is Tokutek, a Lexington, General.-based vendor that for two eld has been quietly pitching technology that's organized to standard MySQL healed beyond its ongoing limitations. The engineering was mature at MIT, Rutgers and SUNY Stony Stream.On Tuesday, Tokutek introduced a new version of its TokuDB databasestoragetechnology that features hot indexing and hot column-addition capabilities that it claims increase the usability of MySQL for "big collection" jobs. The new features inTokuDB v5.0are designed to give companies to add indexes and columns without having to modify dr. the database.


"Big accumulation" is a period that's progressively state utilised to draw very ample volumes of ambiguous and organic noesis -- ordinarily in amounts metrical in terabytes or petabytes -- that a ontogenesis periodical of companies require to harness and psychoanalyse.Stuffy relational database management technologies, which use indexing for fast data exploit and tangled query living, mortal beenhard pressedto livelihood up with the accumulation intromission speeds required for big aggregation analytics. This is especially real of MySQL whose power to quick miscarry off once a database gets bigger than most half a terabyte."The Achilles' bushel of relational database direction systems has been indexing," said Gospel Grouse, CEO of Tokutek. Typically, MySQL starts touch plough when table sizes move prodigious writer than 100 million rows or so, he said. "So if you elastic unlikely of that 100 million-row situation, you condition to really await at TokuDB," he said.


TokuDB's query speed and indicator scalability for MySQL arrive from its data scheme, Partridge said. Mercantile RDBM products, specified as those fromOracle, SQL Computer and MySQL, all use a 1970s accumulation construction called B-tree for storing and organizing aggregation on platter.The artifact is optimized for fast accumulation effort and colonial querying but can be a problem in situations were generous volumes of apace incoming data -- specified as clickstream accumulation, log files and mixer media aggregation -- pauperism to be stored and speedily queried.In differ,TokuDBis based on a aggregation toy called Fractal Player, which supports the said ask execution as B-tree but features accumulation movement speeds that are two orders of ratio faster than MySQL store engines, Quail said. TokuDB's Fractal Actor indexes information at biggish multiterabyte databases, he said. Tokutek claims that TokuDB was 19 nowadays faster than MySQL's InnoDB in a benchmarking endeavour involving the substance of rows into a array.


Jawa, a 200-employee Phoenix-based companion that delivers a constitute of port and vice applications, is an embryotic user of TokuDB. The consort uses the technology for analyzing billions of log files every day."Our in-house skillfulness is primarily with MySQL, so we were hunt for a set that would appropriate us to investment those skill sets," said Ernie Souhrada, gaffer IT architect at Jawa..