Is SQL Server Mature?

  • My first job out of college was developing in PHP against a MySQL DB running on FreeBSD. I actually think it was a good solution and easy to work with given that we couldn't spend a penny on licensing. At least now MS has the BizSpark licensing which makes it affordable for startups.

    Aigle de Guerre!

  • Markus (9/26/2014)

    ...

    Don't look now Microsoft... MySQL is a low end server RDBMS.

    Last time I looked it wasn't, it was a simple file system storage system (with SQL). It's the new xBase, not the new SQL Server.

    I'm a DBA.
    I'm not paid to solve problems. I'm paid to prevent them.

  • At least now MS has the BizSpark licensing which makes it affordable for startups.

    Not when you factor in the tax costs of the licenses. And Microsoft doesn't hide that fact, but they won't help you calculate it either. If you aren't developing an mostly Microsoft solution, it's usually more cost effective to develop using OSS.

  • My company takes a middle of the road approach. We don't upgrade every version but we do try to stay in a version that is supported. When a version is about to go out of support we upgrade to the latest available version, which usually skips more than 1 in between.

    I'm not sure the "If it isn't broke, don't fix it" policy would work for us. What does one do if it does break after it's out of support?

  • Markus (9/26/2014)


    Pricing... yea... Microsoft is starting to think they are Oracle.... soon they will be pricing themselves out of the small applications space. There are starting to be a lot of alternatives that are way cheaper now. Don't look now Microsoft... MySQL is a low end server RDBMS.

    I've been starting to get into Oracle lately (my company runs our reporting out of an Oracle database), and it seems to me that many of the new SQL Server features are often just Microsoft's version of Oracle (and maybe other RDBMS) features. I recently took an Oracle training class on 11gR2 and was surprised to see in memory tables available. So the new feature that is the main selling point of SQL Server 2014 has been around in Oracle for at least 5 years! I don't want to start an Oracle vs SQL Server debate here, but if Microsoft wants me to pay that much more to use the newer version, they need to come up with more than another company's old ideas.


    [font="Tahoma"]Personal blog relating fishing to database administration:[/font]

    [font="Comic Sans MS"]https://davegugg.wordpress.com[/url]/[/font]

  • This is SQL Server: you might not be able to get MS support but if you have a db inconsistency or something chances are their are 3rd party guys (even say IBM) willing to help. Heck I had problems eons ago when migrating a Sun box to a new IBM san and couldn't get the FC card configured properly to see the LUNs. Both Sun and IBM engineers on the line working through the problem, 3rd party db vendor, local IT, and me (programmer analyst at the time representing the user department) on site. Ended up being the IBM guy that figured it out: so what did having vendor support help? There is always someone willing to help for $300/hr even for old hardware/software or equipment that they didn't sell.

    I think the bigger thing is the potential liability though: say you get hacked and customer CC data is stolen. It then comes out you weren't doing "proper maintenance" of your IT systems.

  • s.bradbury (9/26/2014)


    I realise this applies to the editorial less than the SQL Server 6.5 example, but 2008 r2 is such a good product I can't see any real reason to upgrade to a newer Version of SQL Server. Naturally that also depends on its usage...

    More and more I think that key to continued adoption and growth of SQL Server is licensing by scale, not edition, and not by feature. Just like Azure, let me pay for the cores and RAM I need, and let me easily grow that as needed.

    <-- I agree with this.

    I'd agree. I wish I'd have bought a few thousand R2 license that I could now sell 😉

  • pjdijkstra (9/26/2014)


    We are in the proces of dropping our Oracle DB + SAP BODS + Cognos stack and starting fresh with SQL Server 2012 (we asked for 2014 but our mother organisation doesn't 'support' it yet 🙁 ). It is a huge price drop for us and our business case was approved very fast.

    If you come from previous SQL Server versions and have seen the price rise multiple times the perception is very different.

    For us it has 'recently' become mature enough to replace our BI stack.

    Well that's interesting. How much, percentage wise, is the cost reduction? Is the price of SQL Server new that much less than support for Oracle? Are you dropping Cognos?

  • Yes, we´re dropping Cognos and i´m very glad about that.

    I don´t know about the people at your company but the first thing every sales or marketing person here does is export to excel

    Cognos is too expensive to be a glorified excel export tool.

    We went from 200K euros to 35K euros replacing the Oracle+SAP BODS+Cognos stack with MS SQL Server 2012 BI stack.

  • Dave62 (9/26/2014)


    My company takes a middle of the road approach. We don't upgrade every version but we do try to stay in a version that is supported. When a version is about to go out of support we upgrade to the latest available version, which usually skips more than 1 in between.

    I'm not sure the "If it isn't broke, don't fix it" policy would work for us. What does one do if it does break after it's out of support?

    First, lots of people use SQL2K/SQL2K5 just fine. It makes me question support costs.

    Second, if it breaks, what does that mean? You may not want to deploy new applications, but if existing ones run, what's to break?

    Certainly security is an issue, and AFAIK, 2005 just came out of security coverage (or well next year). At that point, I think you need a plan to move quickly if something is disclosed, but otherwise I'd run.

    If there are potentially security issues, you can also think about locking the system down with additional firewalls and other limitations if your application can manage around those.

  • Steve Jones - SSC Editor (9/25/2014)


    Comments posted to this topic are about the item <A HREF="/articles/Editorial/116524/">Is SQL Server Mature?</A>

    That question reminds me of the ads where the scientists are trying to figure out if the cheese is ready or not ready. 🙂

    Seriously, though, I think SQL Server is fairly mature as a database engine and pretty much also with regard to its other components. But I think the Reporting Services piece still feels like it's not quite as mature, especially the report design and viewing. I'm still working on mastering that, so maybe I just don't know the reporting ropes yet, but IMHO that is the part that feels palpably behind the other SQL Server offerings.

    - webrunner

    -------------------
    A SQL query walks into a bar and sees two tables. He walks up to them and asks, "Can I join you?"
    Ref.: http://tkyte.blogspot.com/2009/02/sql-joke.html

  • But I think the Reporting Services piece still feels like it's not quite as mature, especially the report design and viewing.

    I've never understood why Microsoft bought Dundas only to abandon everything good about it. Dundas was a really great plugin for SSRS and allowed the UX challenged like myself to be able to generate decent looking reports. It also was much more user friendly and had several features missing in SSRS. MS bought them and incorporated nothing of their suite. Tragic.

    Aigle de Guerre!

  • Is SQL Server (the core product) mature?

    Well, it has some of the hallmarks of a mature product - very limited feature growth (bad point: very limited bug fixing of even ancient Connect items), no more major changes, etc. Security vulnerability fixes are moderately uncommon.

    The complexity of pricing, however, is awful. What do we need for this instance? Well, we have a 1 core VM, but you can't buy 1 core, they come in sets of 2. But you can't buy just two, you have to buy four! But it's an upgrade from 2005, with/without SA, where it had a processor (socket) license! And so on and so forth, with it getting more complex in 2012 and 2014 both with the different rules on what is and is not allowed. I fear that MS licensing has become or is become more complex and difficult than Oracle licensing!

    The price itself is going up a lot, and yet it's obvious MS is not investing in the (core) product in ways that benefit, well, me at least, and probably many of you. Contained databases? Sure, that may be useful for the Azure cloud, but I have no use for them. Hekaton? Cute, but without FK and CK constraints available, I find it an extremely limited use case for a properly normalized database where data integrity is important. Delayed durability - are you kidding, if I didn't want ACID compliance, I wouldn't be using SQL Server!

  • Leaving price out of the equation, SQL Server is mature at the core. Various early add-ons are mature as well, but not all.

    Microsoft reporting services has made major progress but is not keeping pace with others in the space. Some have taken steps to include within the database the entire line of BI tools fully integrated with the base reporting functionality. Microsoft has not. This may have been a tactical error, time will tell. But is SQL Server a mature BI product? Only in part, Excel is maturing while SQL BI does not seem to make much progress.

    For the most part SQL is the answer. The basics are great, tables, indicies, queries, linked server, views and other mainstays are rock solid. People do not like some of how it was or is done, but different people have different opinions no matter what.

    In my opinion, SQL is mature, and is maturing further with each release.

    Happy Friday!

    M.

    Not all gray hairs are Dinosaurs!

  • Maturity is a subjective concept. It has gone more than 3 versions then by standards and the fact it is often adopted shows it is mature. The question always has to be is it the most effective tool for my needs, and my answer is yes (for my needs).

    So then it is a matter of weighing cost. The idea of an orginization using an outdated version so significantly old is really a concern, especially for matters for security of company, clients and they information they store. Most companies get in the rut of purchasing software and when it comes time to update they can easily skip 1 version, but skipping 2 or 3 is a questionable practice as this shows they are willing to sacrifice staying up to date which has hurt a great many corporations and many data breaches have been with out of date software. Also what happens say in the example if the system get's corrupted and cannot be recovered and they have to purchase something, are they prepared to absorb the cost on the spot, and if so why not reduce their exposure rather than putting off.

    Too many companies in my opinion try to hard to cut overhead and don't really weigh their risks. If the system was storing a list of unimportant pieces of info that if lost or the app could not be recovered that is one thing. But mission critical and security critical items should be on a regular refresh, patch, and recovery testing cycle or you can ultimately expect somwhere you will pay a bigger price tag for lax concern.

Viewing 15 posts - 16 through 30 (of 41 total)

You must be logged in to reply to this topic. Login to reply