If you have never heard of the name before, you are pardoned. Microsoft has not been very vocal about this while its team was working on Project Hekaton. Until now, that is.
Wacky name aside, this is Microsoft’s ambitious solution to bring some lethal speed boost to SQL Server. Microsoft SQL Server databases, like pretty much all other software, rely on traditional hard disks. And while they are fast enough for most applications, some processes could do with even more speed.
Processes like online transactions for instance, which require lots of reads and writes.
But if one is to take the physical layers out of these processes, say taken to the memory and handled there, the whole process could be sped up manifolds.
How many folds you ask? How about even 50 times faster?
Apart from such an intense increase in speed, doing this would also prevent extra wear and tear to the hard disks, which is always a good thing. And besides, it would save enterprises, large and small, from the performance drags associated with managing large data.
Other provider like SAP and Oracle are already using this in-memory technology for speeding up their processes, with solutions like HANA and Exadata, respectively. These are expensive solutions, after all, meaning only big enterprises were using SAP’s and Oracle’s solutions.
Now Microsoft is set to change that by bringing this for its SQL customers. The team is testing the solution, and shared information on Hekaton on Microsoft’s Research website here. As David Lomet, the principal researcher and manager of the Database Group at Microsoft Research, explains:
“There are several in-memory database systems on the market, but what really sets Hekaton apart is that it will be integrated into SQL Server as part of Microsoft’s suite of xVelocity in-memory technologies currently available in SQL Server 2012. Customers won’t need to buy and manage a separate product.”
So essentially not only is Microsoft developing this cutting edge technology, the company is gearing up to integrate it with SQL Server, instead of offering another package. I am game.
Only minor drawback for now is that the database can only run in memory on a single server. But there is no upper limit to how much memory a single database server can use, and with memory prices at an all-time low, that isn’t much of a drawback anyway.
Regardless, big data processing is fast becoming a reality for medium sized businesses, and Hekaton will most definitely appeal to database administrators in the need for speed or massive data transactions.