SQLServerCentral / Discuss Content Posted by Brian Knight / Article Discussions / Article Discussions by Author / SQL Server Needs Incremental Updates / Latest Posts

RE: SQL Server Needs Incremental Updates

David Benoit — Fri, 29 Apr 2005 10:41:00 GMT

The length of time between releases is not a major thing with me or my organization especially if you have a product that works and SQL Server 2000 does just that. The big constraint that I totally agree with Brian on is the significance of the changes, which are needed and are extremely exciting, BUT are huge and as a result are potentially overwhelming to DBA's (explain below) and to the organization in the implementation process. I have been working with the BETA since the first release and some areas have taken me a LOT of getting used to. Without going on another rant, I have been poking at MS for quite some time (not through effective channels obviously) that they needed to develop a better training set so that they could increase the skill set of the SQL Server DBA's in the technology world today. I believe this release is going to fully expose this weakness. The SQL Server 2000 product set is so easy to configure and maintain that it has bred a weak group of DBA's in the market (I say this with fear in that I know there are a lot of great ones too, most on this site, covering up here....). This next release of SQL Server changes things significantly. I am enthusiastic about the Tech Presentations that are being run and I am working with the DBA's on my team to work through these but again, this is not going to be enough AND I really don't believe that the standard training offerings are going to be sufficient to bring this release of the DBMS to the level of support that is going to be required in the larger organizations where millions of dollars of business are going to run through this platform.

So, couple big changes with a significant increase in skill set required to maintain the product and I believe you have a big potential for a problem in relation to your products appearance to the customer. I'm highly confident that the product will be solid upon release and will be an extremely efficient DBMS but to make the product perform to the level that is going to be required for your larger organizations, which is where they are trying to go today with this release, is going to take a different mentality and breed of DBA's. Not sure this is being taken into account as it should....

Ok, I know, I ranted.... Sorry.

RE: SQL Server Needs Incremental Updates

fewgoodpeople — Sat, 25 Sep 2004 09:13:00 GMT

I can't agree with Brian more. Without native XML storage capability, and a major release in 5+ years, SQL Server does compromise its market position. The recommendation about VS IDE integration of yours was insightful and some gradual changes to DTS are neccessary. Looking at the DTS 2005 now, I realize I need to unlearn a lot and learn a lot.

The newly added For XML Path switch in SQLXML (TSQL) is wonderful, my SQL Wish to this would be just that SQL Server can generate a template for the predictible SQL Scripts (For XML Path) once the users tell SQL Server the XSD required. This is particularly convinient when we have a 200+ elements of XSD that we need to fulfill.

After this long wait, hopefully we can have a SQL Server 2008 ready!!!!

RE: SQL Server Needs Incremental Updates

Michael LeVere — Wed, 05 May 2004 08:04:00 GMT

David,

I totally agree with you about Sybase being primeval ooze. With the upgrade to Sybase 12.5.1 they just added scheduling services for jobs. Something that SQL Server has had for the last 5-7 years. I have been working with Sybase for the last 4 years and I have always thought that Sybase was lacking. What it isn't lacking is speed. Running on Sun Solaris 9 we are able to run the 64 bit version of Sybase with great speed. We are also able to take advantage of 24Gb of ram for caching. Compressing dump files without having to go through a third party application as well.

There are alway going to be give and takes for both of these applications. If we can get to the 64 bit version of SQL and Win2003 we would gladly switch over.

RE: SQL Server Needs Incremental Updates

David.Poole — Fri, 30 Apr 2004 07:46:00 GMT

I would be curious to know where Sybase ranks along side SQL Server, ORACLE, DB2 etc.

I have colleagues who are using Sybase on a particular project and because both SQL Server and SyBase come from the same parentage I get asked a lot of questions.

There are a number of times when my Sybase colleagues have to say "I'm sorry but we don't have that feature within Sybase".

The over-riding impression I have gained from these comments is that if SQL2000 is jurassic then Sybase is still primeval ooze.

Would anyone care to comment on this?

RE: SQL Server Needs Incremental Updates

Frank Buchan — Fri, 30 Apr 2004 07:39:00 GMT

Not having had time to dig into all the raw details of all the available information, I may be mistaken, but it does seem that they will still lack some low-level control over indices and that their analysis tools will still be limiting in some regards as to the management of analysis cubes, etc. I admit to some ignorance on the correctness of those specific feature criticisms though, which have come to me indirectly via database administrators whose experience with recent versions of Oracle and DB2 have led them to conclude that these features are lacking. While I trust their opinions, it in no way makes them correct.

Probably, in retrospect, I should have removed the word "notable" from that previous post, since I'm not sure that even giving merit to these critiques makes them "notable" in any way. Speaking solely from my perspective, I suspect SQl Server 2005 will be feature-rich in a way that take sme many month to even dream up an implementation that taxes it.

I suppose some days the light bulb in my head is brighetr than others.

RE: SQL Server Needs Incremental Updates

David.Poole — Fri, 30 Apr 2004 01:43:00 GMT

Could you give examples of the "notable exceptions"?

RE: SQL Server Needs Incremental Updates

marris — Thu, 29 Apr 2004 21:13:00 GMT

I agree that in order to compete with other DBMS vendors MS needs to come out with releases often. Five years is too long a time especially when SQL Server is lacking in security features. But on the other hand constantly releasing new versions mean that companies need to upgrade constantly which is an inconvenience when the older version becomes unsupported by MS.

RE: SQL Server Needs Incremental Updates

Sami-30112 — Thu, 29 Apr 2004 17:01:00 GMT

I agree with Brian. In my case, it will take at least 6 months to test, wait for a service pack, test again and then get management buy in. By that time, it will end up being more of a 2006 implementation.

We were able to upgrade from SQL 7 to 2000 in a very shore time frame due to the smaller amount of changes. I just don't see how that would be possible here, unless this is one of the few times when Microsoft get's it right the first time. Their track record on first releases is not good.

As Steve Brett said: rant over

RE: SQL Server Needs Incremental Updates

Mark Whitfield — Thu, 29 Apr 2004 12:27:00 GMT

I think Brian has the right approach, but for a different reason. Part of the process of an upgrade is the testing that MUST occur for all of your applications that use the database. Testing against the huge set of features in the SQL Server 2005 release is going to be a mammoth task that, frankly, most companies are going to gloss over or skip altogther. Having a smaller feature set allows that testing to be much more concise and manageable by an already downsized and overworked database admin and development staff.

RE: SQL Server Needs Incremental Updates

SteveB-74825 — Thu, 29 Apr 2004 09:55:00 GMT

Inded.

Although i WAS on the beta and am now not. I've moved companies.

Ironically the company i worked for were really lucky to be on the ascend program and they're not interested in yukon at all. they'll probably deploy it but are 'too busy' to follow the beta process though, which is a shame when there are companies out there that that would be interested.

Rant over. My boss was a nice guy. Really he was.

RE: SQL Server Needs Incremental Updates

Antares686 — Thu, 29 Apr 2004 06:57:00 GMT

Personally I think they should put the beta out for the public to download, or request since there are some many new features. The only people who will be ready are those lucky enough to be on the beta. I agree too much to absorb in one go and even when released it will be a year before most actively consider. A public available beta will allow many to adopt faster with more confidence.

RE: SQL Server Needs Incremental Updates

Frank Buchan — Thu, 29 Apr 2004 06:56:00 GMT

I believe that part of the situation MS has faced is that SQL Server has been playing catch up for a long time. Even though I use SQL Server 2000 extensively now in my day-to-day development, its weaknesses have been very evident. Some might have been addressed in service releases, while others were of a nature that would have made that difficult.

Looking at the feature set of SQL Server 2005 suggests that with a few notable exceptions they have caught up to the marketplace, and I suspect that after this release, we will see more of the missing elements provided via service releases on a tighter schedule. I am only guessing, but I believe that what took so long with this release was that some fundamental changes had to be made to the core of the server to open that option.

It is also important to remember, though, that they did release some major feature extensions during the wait (such as Reporting Services and the XML Services packages). What has been lacking has been an integrated release that made some of these disparate parts accessible.

Let's hope that with this coming release it is done right.

RE: SQL Server Needs Incremental Updates

EdVassie — Thu, 29 Apr 2004 04:04:00 GMT

I completely agree with Brian.

There are a number of issues here...

1) Comparisom with other vendor's release stategy.

My main experience is with DB2. When IBM are near to making a new DB2 release RTM (GA in IBM terminology), they will make known to technical conferences the major line items in the subsequent release (Vnext), and those line items that will not be considered until after Vnext is RTM. This gives about a 6-year plan for the product's future. They also have about a 2.5 - 3.5 year release cycle and regular (approx quarterly) service packs. The shorter release cycle means an easier upgrade path, with the +1 release supporting everything that has not been deprecated in the current release.

All this allows sites to prepare far more for what is coming, and the regular SPs help greatly in budgeting manpower for maintenance.

2) Credibility of vendor

When a vendor makes it known they hope to release a major update in 3 years, and it actually takes 5 (e.g. SQL Server 2005), you start to wonder about their ability to deliver for the future.

If Microsoft continue to deliver major upgrades a year or two after making their initial plans known, it will have a corrosive effect on their market share, regardless of final product quality.

3) Size of feature set

SQL2005 is massive compared to SQL2000, but SQL2000 is not the main competition. Compared to its main rivals, SQL2005 will deliver VERY little that will not already be in other products.

For SQL2000 people, the feature set size is a problem, especially for DTS. I think it is a major error on Microsoft's part the SQL2005 DTS does not have source compatibility with SQL2000 DTS. Run time compatibility only is simply not good enough for adjacent releases. We will have to re-engineer all our DTS at the time of migrating to SQL2005, to ensure that any further troubleshooting or maintenance of a single DTS package will be compatible wih the other packages it works with. This will add perhaps UKP 200,000 to our SQL2005 upgrade costs, due to the amount of DTS we have.

4) The SQL2005 legacy

The frightening aspect of a 5 year wait for SQL2005 is the prospect of a further 5 year gap for the next release.

In a number of areas, SQL2005 technology will be inferior(*), and while its rivals move further ahead with their 3-year release cycles we will be stuck with what we have until maybe 2010. By then, we will either have a release that makes SQL2005 seem minor, or SQL Server will remain permanently and visibly many years behind the times.

(*) Areas where SQL2005 is inferior include continued index locking for insert/delete activity, single dimension only index clustering, slow adoption of SQL standards changes, etc.

RE: SQL Server Needs Incremental Updates

SteveB-74825 — Thu, 29 Apr 2004 03:28:00 GMT

The learning curve will be steep, but I must admit this is the first time Microsoft has involved developers from ISVs from all over the world in beta testing and design input of any product (so they said !!)

I was on the Ascend Phase 1 training program (Developmentor run) in in Germany in Jan of this year and Yukon will be a mjor change, but it's a major change for the better and I really don't think it could be done in stages. Saying this the new veriosn of Reporting Services was available as a stand alone beta called Rosetta. It's now part of Yukon.

If I was a database developer or DBA I'd spend the next year learning C# and .Net and becoming more familiar with XML. TSQL will still be the language of choice for set based querying but I think C# will become more prevalant as it's a much richer environment to work in.

RE: SQL Server Needs Incremental Updates

David.Poole — Thu, 29 Apr 2004 01:20:00 GMT

I don't like the idea of a huge and radical change to something as fundamental as the SQL.

Don't get me wrong, I'm enthusiastic about SQL2005 its just that the learning curve, from what has been said so far is going to be pretty steep.

That said a release of a database tends to be pretty major, and therefore it is not something you want to be doing every year.

I like the way that Microsoft have released the Reporting Services as a bolt on to SQL2000.

It would be good if the various components of SQL Server were available as bolt-ons so if a huge upgrade occurs you simply upgrade the bolt-ons as and when is appropriate.

RE: SQL Server Needs Incremental Updates

Steve Jones - Editor — Sun, 25 Apr 2004 10:21:00 GMT

A good idea, but I'm not sure I agree. Unless there is some pick and choose and a subscription model. Upgrading our DBs is a major challenge, both in time and resources (testing) and politically (money), so having the releases fewer and father between is something i like. If Yukon released in Q3 or 4 of this 2003, there's no way we'd upgrade for a couple years. However with the release next year, it's an easier sell to start testing.

SQL Server Needs Incremental Updates

Brian Knight — Sat, 24 Apr 2004 15:23:00 GMT

Comments posted to this topic are about the content posted at http://www.sqlservercentral.com/columnists/bknight/s