VectorWise has emerged from stealth mode. Even more interesting is that Ingres is collaborating with them on Ingres VectorWise. And most interesting is the video made to promote the work. In the interest of disclosure, I have friends who work on this at Ingres and VectorWise.
I have read many of the papers on X100 and MonetDB and believe the performance claims they make. Within the MySQL community, this is most similar to Kickfire in terms of the query workloads it can support. The difference is that VectorWise writes their software to run at full speed on a modern CPU and maintain a high IPC rate while Kickfire uses custom hardware (and a lot of really good software that they probably cannot describe without an NDA). Anyone attempting to build a VectorWise storage engine would have to do all of the fun things that KickFire has done to circumvent much of the MySQL optimizer and execution code, so I doubt that will be repeated.
Wednesday, July 29, 2009
Subscribe to:
Post Comments (Atom)
Posts
Hi, Mark, thanks for a nice summary. Optimization for VectorWise is done by improving and extending the existing optimizer. It wasn't clear at the start how difficult this would be, but with some creativity from Ingres and VectorWise the VectorWise plans are as good and sometimes better than the hand-written x100 plans, and some of the improvements benefit also to the existing Ingres storage formats.
ReplyDeleteall this stuff is very nice but I waste at least 2 hours trying to find in web some real evidence of their performance result or may be monetdb with x100 inside, just to try and even got full (last) Ingres svn ))) before it's possible to reproduce some of their results all this big piece of of shit. I have seen working KickFire solution and fully understand how and why is it works. If you can provide some working piece of code (or executable, although these guys are "open source" ))))
ReplyDeleteHi Anonymous,
ReplyDeletejust a few days after your post (which I did not notice), the Ingres VectorWise preview program started. You can register for beta and download/evaluate now by going to www.ingres.com/vectorwise
..and it won't be long before this extremely powerful database system is out of beta.
Peter Boncz