The Decline of SQL Server

  • EXACLTY!

    They're getting greedy, and it's going to hurt their market share.

    Here's to hoping they quickly realize their mistake.

  • Can anyone recommend a wiki for MySQL?

    Is there a good wiki for SQL Server?

  • lsmith 90637 (10/3/2014)


    Of all the OS and Service product lines that they've ever fielded, SQL Server has been THE consistent winner. Even when every other release of Windows falters, SQL Server rolls on.

    Cherry picking quotes here, but this quote sums up what how I feel.

    IMHO, the OS has always been SQL Servers weak spot.

    I remember using 2005 and thinking it was OK for small to medium stuff but fell flat on its face in a VLDB environment. 2008 Comes around and I feel much the same. The DB seems more robust, but the underlying OS just can't handle (in lots of testing at the company I was with at the time) 4 9's uptime like our *nix servers could...even running Oracle.

    Now, 2012 looks even better than 2008, but I still have serious reservations about the OS.

    I know this will be heresy on this board, but I would love to see a SQL Server version for *nix.

  • In the company I work for MS has been a de facto standard for at least 15 years. Now, we are starting to see Oracle being tossed around as an option. And MS has no one to blame but themselves: this is a result of the unseeming pricing changes it implemented.

  • SQL Server is a great product, but the licensing cost of recent versions has forced us to loo at what else is on the market.

    Amazon Redshift provides much the same column-store and in-memory capability as SQL 2014 Enterprise Edition, but at roughly Standard Edition costs. We are now migrating our BI to Redshift as the cost differential makes this a worthwhile move.

    We have been heavy users of SQL Full Text search, but there has been no improvements to this since SQL 2008. Semantic Search looked promising, but is restricted to returning a maximum of 10 matches per query which is at least 40 too few for us to consider using it. We have had to look outside of SQL Server to improve our search technology.

    The end result is that from around 50 SQL Server instances in 2011 we are moving to maybe 9 instances in 2015. Some of this reduction is due to consolidation on modern large servers, but there is still a big reduction in number of cores licensed.

    On a recruitment basis we still ask mainly for SQL Server skills, so our profile in the jobs market does not (yet) reflect the reduction in SQL Server use but I expect this will change.

    SQL Server used to be the most cost-effective solution to many of our production data needs, but there are now many mature products each able to take a bite out of the SQL Server lunch. I hope the next version of SQL Server will address the cost problems, as there is a risk that by the time a trend is seen Microsoft will have already lost a significant amount of market share.

    Original author: https://github.com/SQL-FineBuild/Common/wiki/ 1-click install and best practice configuration of SQL Server 2019, 2017 2016, 2014, 2012, 2008 R2, 2008 and 2005.

    When I give food to the poor they call me a saint. When I ask why they are poor they call me a communist - Archbishop Hélder Câmara

  • Some of these sentiments remind me of what I heard (i.e. total hearsay) after IBM took a big hit (late 70s/early 80s?): IBM upped their prices in the false belief that no-one would select an alternative even when they were being told otherwise by both their (ex-)clients and sales figures. In the end you can argue that they only just survived. Such arrogance!!!

    Gaz

    -- Stop your grinnin' and drop your linen...they're everywhere!!!

  • My thought is that MS is raising prices and at the same time changing their target audience with SQL Server. Their focus seems to be shifting to the big data market and away from the mid to small size customers because they've hit a saturation point. My company is shifting to Teradata for some of the larger systems because their is still a belief that MSSQL isn't scalable, but we'll still keep MSSQL for smaller systems.

    Aigle de Guerre!

  • Meow Now (10/6/2014)


    My thought is that MS is raising prices and at the same time changing their target audience with SQL Server. Their focus seems to be shifting to the big data market and away from the mid to small size customers because they've hit a saturation point. My company is shifting to Teradata for some of the larger systems because their is still a belief that MSSQL isn't scalable, but we'll still keep MSSQL for smaller systems.

    ...then they should expand the range of editions available. Keep lower editions price competitive (from the free SQL Server Express) and raise the price of the high end SQL Server editions. Allow for mix'n'match of editions on sites and provide upgrade and (optionally) downgrade pricing.

    Gaz

    -- Stop your grinnin' and drop your linen...they're everywhere!!!

  • We're tied tightly to Microsoft in our company.

  • So there are 10 startups who don't use SQL Server or Oracle to every corporate client. They probably don't even hire a DBA. Most of their development is with an ORM, so their devs don't rely on SQL. Do you consider market share by the company, server, instance, number of positions with technical expertise needed?

    We use several 3rd-party application providers that all use SQL Server almost exclusively. It is too costly for them or us to switch.

  • A *nix version of Sql Server might be what's currently offered in Azure. If it isn't there already it soon will be in order to handle the demands of cloud services. I think converting some of the servers over to Postgres would be great except that the developers will have some challenges to overcome. It's good business sense to save $100k-$300k in licensing.

    The price increase is a way of forcing clients into the cloud. Either get good at making sound business decisions on-premise or offload as much of those services onto a distributed computing (CLOUD) vendor ASAP. Economics will play a huge role in the viability of several businesses in the next 5-10 years.

  • Gary Varga (10/3/2014)


    jay-h (10/3/2014)


    chrisn-585491 (10/3/2014)


    One word: Java

    Horrendous mess.

    But you know what? They delivered on the cross platform problem. Nothing else installs on OS X, Windows and Linux with as few issues. JetBrains can create some decent IDEs that work well on all platforms. Pentaho can work better than SSIS in many cases. And Clojure may bring Lisp to the masses.

    I'm not sure about that 'cross platform' except in advertising copy. We've got major apps that will only run on certain VERSIONS (forget platforms) of Java and won't even coexist on a machine. And it's not uncommon for a security patch to seriouslyl brake things.

    I think that the joke was: "Write once. Test everywhere."

    As a Brit I prefer Write Once Debug Everywhere - WODE

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

  • andrew gothard (10/9/2014)


    Gary Varga (10/3/2014)


    jay-h (10/3/2014)


    chrisn-585491 (10/3/2014)


    One word: Java

    Horrendous mess.

    But you know what? They delivered on the cross platform problem. Nothing else installs on OS X, Windows and Linux with as few issues. JetBrains can create some decent IDEs that work well on all platforms. Pentaho can work better than SSIS in many cases. And Clojure may bring Lisp to the masses.

    I'm not sure about that 'cross platform' except in advertising copy. We've got major apps that will only run on certain VERSIONS (forget platforms) of Java and won't even coexist on a machine. And it's not uncommon for a security patch to seriouslyl brake things.

    I think that the joke was: "Write once. Test everywhere."

    As a Brit I prefer Write Once Debug Everywhere - WODE

    As another Brit (half English / half Hungarian), I prefer C# 😀

    Gaz

    -- Stop your grinnin' and drop your linen...they're everywhere!!!

  • Gary Varga (10/9/2014)


    andrew gothard (10/9/2014)


    Gary Varga (10/3/2014)


    jay-h (10/3/2014)


    chrisn-585491 (10/3/2014)


    One word: Java

    Horrendous mess.

    But you know what? They delivered on the cross platform problem. Nothing else installs on OS X, Windows and Linux with as few issues. JetBrains can create some decent IDEs that work well on all platforms. Pentaho can work better than SSIS in many cases. And Clojure may bring Lisp to the masses.

    I'm not sure about that 'cross platform' except in advertising copy. We've got major apps that will only run on certain VERSIONS (forget platforms) of Java and won't even coexist on a machine. And it's not uncommon for a security patch to seriouslyl brake things.

    I think that the joke was: "Write once. Test everywhere."

    As a Brit I prefer Write Once Debug Everywhere - WODE

    As another Brit (half English / half Hungarian), I prefer C# 😀

    sounds like the dirty little secrets of Java are coming out. 'write once test everywhere!'

    As I recall, going way back, Microsoft stopped supporting multiple platforms after it had issues with its spreadsheet tool Multiplan (it was out before lotus 1-2-3, but after Visicalc)

    The more you are prepared, the less you need it.

  • Is that I reference to Xenix I hear? I once saw a Xenix box at a place I worked but never saw it running.

    Gaz

    -- Stop your grinnin' and drop your linen...they're everywhere!!!

Viewing 15 posts - 46 through 60 (of 64 total)

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