Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Joe NYC who wrote (113106)	10/11/2000 12:08:10 AM
From: rudedog	Read Replies (1) \| Respond to of 186894

IBM's DB2 has excellent partitioning and parallel query capability, as shown by their TPC result earlier this year. The latest version SQL Server has good partition capability and reasonable parallel capability but not as good as DB2. Oracle's design is not well optimized for either parallel queries or partitioned data, as it was originally built around a shared disk model (based on the old VAX Cluster architecture) and therefore requires a distributed lock manager, which is a high overhead bottleneck. That's one of the reasons Oracle has not been prominent in recent TPC performance leadership. Oracle's design does pretty well in a large SMP environment where the lock overhead is much lower, but even there, they have fallen victim to more modern designs. Take a look at tpc.org and you will see that the only Oracle listing is #10 - the CPQ Alpha result, and I attribute that to the very strong performance of the Alpha processors, and the DEC heritage in the Oracle design. MS SQL owns all of the top 10 in price / performance, see tpc.org

To: Joe NYC who wrote (113106)	10/11/2000 12:13:38 AM
From: rudedog	Read Replies (1) \| Respond to of 186894

Jozef - in answer to your question about transparency - Existing database access programs will run against a partitioned database with no changes - but they will not take good advantage of either the partitionability or the parallel query capability. Modest changes in table structure and indexing can improve performance without major surgery. The biggest gains come when the data schema is laid out with the partioning structure in mind. So in some ways it is like the transition from 32 bit to 64 bit - 32 bit code will run, but the real benefit comes from applications designed for the 64 bit world.