October 17, 2008
Oracle notes
I spent about six hours at Oracle today — talking with Andy Mendelsohn, Ray Roccaforte, Juan Loaiza, Cetin Ozbutun, et al. — and plan to write more later. For now, let me pass along a few quick comments.
- The key philosophical point that I had perhaps been missing is that Oracle thinks there is and should be a storage (server) tier, just as there also are database (server), application (server), and web (server) tiers.
- Exadata cells are designed to never talk with each other. Instead, they talk to a set of Infiniband switches, which then talk to a grid of servers on the database tier. Oracle thinks this has solved its I/O bandwidth problem for once and for all. It’s hard to see why that wouldn’t be the case.
- What Exadata does on the storage tier in query execution is throw stuff away. Mainly, this is projection and restriction/SELECT. But if a join has been resolved on a small fact table, and Oracle is now filtering a fact table to match a value or set of values, the storage tier can do that too.
- Backups are now done (or soon will be?) on the storage tier. I presume the same goes for restore, but I didn’t ask.
- Oracle says that RAC (Real Application Clusters) no longer has much to do with locking (RAC’s antecedents include Distributed Lock Manager, which is why the question arises). Generally, Oracle denies that RAC creates overhead or bottleneck problems.
- Generic UDFs (User-Defined Functions) aren’t automatically parallel on Oracle, unless it’s clear that they only affect single rows EDIT: or otherwise are assured to be free of side effects. However, Oracle has itself implemented and shipped long lists of parallel UDFs.
- Oracle insists that even without Exadata, it has lots of users in the 10+ terabyte range, and few over 100 terabytes. Oracle further insists that this is with the most conservative kind of counting — single database, true user data, etc. Oracle estimates that Teradata Petabyte Power Player Dell would only be at 300 terabytes by this kind of counting.
- I didn’t focus on admittedly worthy questions like “Across how many cores do these parallel analytics really scale?” or “How easy is it to provision a new node in a RAC cluster if one goes down?”
- The SAS drive option has been increased from 300 to 450 gigabyte drives. Presumably, this will take our estimate of high-end Exadata list pricing down from $198K/TB of user data to $122K. Competitive vendors should show similar improvements, however, if they also use new generation drives.
Categories: Data warehousing, Exadata, Oracle, Parallelization, Pricing, Storage, Theory and architecture
Subscribe to our complete feed!
Comments
10 Responses to “Oracle notes”
Leave a Reply
Real Application Clusters and Oracle Parallel Server alike require a Distributed Lock Manager. Since Oracle8 Parallel Server, the DLM has been provided by Oracle. Prior to that it was provided by the host platform vendors.
Thanks for the clarification Kevin … but what is Oracle Parallel Server? 🙂 Simply the predecessor to RAC?
PS. In other news, you have an is/his typo that needs fixing in a short recent post on your blog.
Hi Curt,
Yes, Oracle Parallel Server (OPS) is in fact a prior release of RAC. It was re-branded when Cache Fusion was implemented (circa Oracle9i). Actually, Oracle8i R3 (8.1.7) had little bits of Cache Fusion in it too (for certain types of cross-instance reads). Trivial pursuit I suppose.
Thanks for the typo tip.
PS. I heard your visit at Oracle HQ went well.
Kevin,
That was my sense too.
An attempt was made to locate you toward the end, whereupon it was discovered that you don’t live in the area or work at HQ. 😉
Based on a couple of remarks, and some follow-up discussion, it seems safe to say that you and I together have helped educate Andy Mendelsohn on the influence of blogs. 🙂
Curt,
That is funny because Juan recruited me and has known me for 18 years (joint engineering with between Sequent and Oracle) and is my bosses-boss…funny.
Juan has bought into the value of blogs. I’m glad to see Andy joining in.
Kevin,
Juan was probably there for the shortest stretch of time, possibly excepting Vishu Krishnamurthy, who was only invited when it became clear there would be time to talk about something besides data warehousing. 🙂 I.e., Juan didn’t sit and stay much except during his actual presentation. So he wasn’t there around the end.
Ray Roccaforte and Gordon Smith are the ones who stayed pretty much all the way through. Andy was in and out. I’d guess Cetin stayed longer overall than Juan did.
Mark Townsend, who was originally tasked with setting up the visit before delegating that to Gordon, wasn’t there at all.
[…] I went to Oracle in October, the main purpose of the visit was to discuss Exadata. And so my initial post based on the visit was focused accordingly. But there were a number of other interesting points I’ve never […]
[…] to it. Obviously, this stands in contrast to the endorsements of Infiniband for data warehousing by Oracle and […]
[…] for data warehousing. At this time, that’s a claim much better supported by marketing and theory than by […]
[…] DB2 perhaps excepted. And I say “last year” so as to duck the questions of whether Exadata finally solved Oracle’s problems and whether Madison will once Microsoft releases it. […]