IBM throws DB2 Power cluster at Ellison's Exadata
Sun thunder stealer
The war of words and technology between IBM and Oracle will get a little warmer today as Big Blue launches its DB2 PureScale clustering technology for its DB2 database and Power Systems Unix servers.
As El Reg reported earlier this week, IBM's database and server techies have been cooking up a clustered DB2 database implementation on IBM's AIX-based Power servers to steal some thunder from Oracle and its minion, Sun Microsystems, at Oracle's OpenWorld customer event next week in San Francisco, California.
Oracle is expected to roll out some sort of Sparc-based Solaris system at the event, most likely a cluster of Sun's T5440 servers running Oracle and Solaris and very likely resembling the Exadata V2 x64-Linux database cluster that the pair announced a month ago.
The basic feeds and speeds of the DB2 PureScale offering are exactly what El Reg had heard through the grapevine, except that unlike the Exadata V2 setup from Oracle and Sun and unlike IBM's own Smart Analytics System for data warehousing, which debuted in late July and which you can get the feeds and speeds of here, DB2 PureScale is not a preconfigured system.
Rather, it is a database clustering feature that is only being made available on AIX running on IBM's Power Systems iron, and specifically only on its midrange Power 550 machines and its high-end Power 595 servers. The Smart Analytics System is a cluster of Power 550 machines tuned to do data warehousing that runs the AIX operating system, the DB2 database, various Cognos data warehousing products, and soon SPSS analytics now that IBM has acquired SPSS for $1.2bn.
Oracle is sure to dig IBM about the fact that DB2 PureScale is not a system, but something that has to be configured (presumably by IBM Global Services), while the Smart Analytics System is a preconfigured box, ready to go to run data warehouses and their analytics software, as is Oracle's Exadata V2.
IBM seems to have gotten the integrated system down with one offering, but not with the other, which is a bit peculiar. According to Bernie Spang, director of product strategy for the information management division of IBM's software group, though, this is just facing up to the reality that one size does not fit all when it comes to servers, storage, and InfiniBand switching when it comes to online transaction processing.
Oracle will no doubt pick on IBM for having one set of stuff for OLTP on database clusters - DB2 PureScale on Power iron running AIX - and another for data warehousing - the Smart Analytics System, which shares many components but which can be quite different.
DB2 PureScale, says IBM, does what many companies have always wanted to do: allow a clustered database to look like it is running on a giant symmetric multiprocessing server, where the clustering is in the chips and chipset and the processor cores share memory. With many parallel database implementations, you have to carve up the datasets and spread them out across database nodes, or you have to tweak your applications so they can run on a parallelized database.
Both IBM with PureScale and Oracle with the Real Application Clusters extensions to the past several iterations of the eponymous Oracle database, say they have this problem licked. Moreover, IBM says that customers will not have to go through a lot of database tuning to make PureScale work. And finally, because it is a clustered database, high availability is built in. One database node goes down, another takes over its work.
The DB2 PureScale feature was co-developed by IBM's database software engineers in its Toronto, Ontario software labs and the Power Systems and AIX development lab in Austin, Texas.
The database feature is being branded with the Power HA (short for high availability) brand, which comes out of the Power Systems division within IBM's Systems and Technology Group. The Power HA software stack includes what used to be called HACMP clustering (now known as Power HA for AIX) and High Availability Storage Manager (now Power HA for i, referring to the proprietary i/OS operating system for Power boxes).
For whatever reason, IBM has not rebranded the iCluster HA clustering product it gained through its $161m acquisition of DataMirror in July 2007 with the Power HA moniker. The important thing that Spang wants people to realize is that the Power HA DB2 PureScale feature - that rolls right off the tongue, eh? - is a new clustering technology and is not based on any of these clustering products.
pureScale has been worked on for over 4 years now
Exadata V1 was likely not quite dreamed up yet when the work on pureScale was started - 4+ years ago - just as scotdb said.
It takes years to build this kind of technology into a product - even when you are borrowing the architecture from another platform.
Oracle RAC is yesterdays technology that was designed on a message passing architecture through sockets - using vast quantities of CPU horsepower to service the context switches and interrupts that such communication entails.
Oracle has gone out of their way to try to get vendors to get emerging technologies like InfiniBand fit into their crappy design by insisting on RDS - so that it looks and feels like sockets - so that they don't have to make the deep investment to exploit RDMA and IB in its native form through uDAPL or MPI - from usespace to userspace - no interrupts - no wasted memcpy into the kernel and system calls etc.
RDMA is here to stay and will be mainstream on 10G ethernet as well shortly. Oracle is behind the curve on this on.
Who do you think invented Parallel Sysplex in 1993? The same company that is introducing it in their Power systems.
Not aimed at Exadata but at RAC !!!
Now that this is officially announced, you'll see that this has nothing to do with Exadata but at those who want to move from RAC to something more scaleable.
And I can assure you that this wasn't "cooked up in the last few weeks" since I was at a briefing on this in June during IOD Europe, and at that point they''d been working on this for a couple of years.
The obvious comparison is with DB2 for z/OS Data Sharing. The underlying architecture and even the terminology being used is the same, and no doubt there is a lot of common code too.
The big benefit of Data Sharing on System Z has always been the Coupling Facility which prevents having to broadcast changes to all the members all of the time. That's where the scaleability comes from. Getting this on AIX is awesome.