By: Philip Howard, Research Director - Data Management, Bloor Research
Published: 9th April 2014
Copyright Bloor Research © 2014
SAP has just released version 16 of what used to be Sybase ASE and is now SAP ASE. There are couple of interesting things about SAP ASE, one of which is related to this release and one of which isn’t. I’ll deal with the latter first.
The acquisition of Sybase by SAP has seen a renaissance for ASE. I’ve mentioned it before but it is worth repeating: Sybase ASE was trailing in market presence behind the 'Big 3' before the company was acquired by SAP and its number of users was more or less static. Since the acquisition, there has been a resurgence of new implementations: I reported last year that these had topped a thousand, and SAP personnel have informed me that there are now over 3,000 SAP Business Suite implementations running on ASE.
What has happened? Is SAP ASE suddenly a much better database? No. Actually, it was a pretty good database in the first place. Of course, it supports SAP application environments, which it didn’t before, and that is the main driver for its resurgence. But you can still implement SAP on Oracle or SQL Server or DB2 and if these are so much better than ASE how come ASE is gaining market share? Of course, dollars may be an issue but isn’t it also market presence? SAP has market presence, Sybase did not—so companies didn’t license ASE before but they do now. Which begs the question: is the IT industry a bunch of lemmings or are IT leaders really looking for best of breed solutions.
Anyway, on to SAP ASE 16. The most interesting thing about this is not so much the features that are being introduced but the direction that SAP is taking ASE in, in this release and also in future releases. And the main focus is big data. Not only in terms of variety but certainly in terms of volume and velocity. As an OLTP database provider, SAP sees ASE customers with increasingly large numbers of transactions to process against appreciably larger databases and with response requirements that are being driven inexorably downwards (I can remember when 2 seconds was regarded as an acceptable response time for end-users but now we are talking about measurements in milliseconds for users and near instantaneous processing [microseconds] for machine-generated transactions).
Now, this is pretty obvious stuff. What’s interesting is that none of the other major OLTP database vendors have talked to me about big data in this context. Of course, maybe they are just not talking to me, period, but I actually think it’s because people have been so focused on big data in the context of warehousing and analytics that they haven’t got around to sorting out their OLTP marketing. So, good for SAP.
As far as the details of ASE 16 are concerned I’m not going to spend a lot of time on this. Besides linear scale up that makes big data for OLTP possible, features are the removal or replacement of query limits (more user tables, support for multiple triggers and more sub-queries), partition level locking, full text auditing, index compression, dynamic thread assignment, various improved query functions (hash joins, sort, star joins), a full database encryption option, and improved and simpler management and administration. In a release later this year, SAP will deeply integrate its market-leading database replication technology with ASE to provide both HA and DR support for custom applications.
The four foci for SAP in this release are scalability, security, speed and simplicity. Which is about right given the new features in SAP ASE 16. I hesitate to suggest this—perhaps I should wash out my mouth with soap—but maybe SAP should start a marketing campaign: “you need the four Ss to address the four Vs (the fourth being veracity) of big data”?
Posted: 13th April 2014 | By Andrew Melkonyan :
Any idea what has been done on releasing the metadata contention in ASE 16? The brief white-paper is pretty cryptic on it...
Posted: 14th April 2014 | By Philip Howard :
The following is the detail I have from SAP:
There are 3 key areas which have been optimized:
1. In the metadata cache - Use count the code for updating use code has been made automatic, and as a result this operation is much faster. Users can choose to use des_bind if desired to bind objects to cache, but this is not needed
2. ELC 50% of the procedure cache is now used for ELC, vs 25% before. This decreases contention on the procedure cache
3. Row size can be set for hot tables, using sp_chgattribute exp_row_size.
Posted: 16th April 2014 | By Andrew Bagley :
Isn't the real story here that SAP is being forced to offer ASE since HANA OLTP performance is not where they would like?
The messages above were all contributed by IT-Director.com readers. Whilst we take care to remove any posts deemed inappropriate, we can take no responsibility for these comments. If you would like a comment removed please contact our editorial team.
All fields must be completed to submit a comment. Email addresses are passed through to the author so they can contact you directly if needed.
Published by: electronicdawn Ltd.