April 11, 2014 at 9:48 pm
How long needs to tempered by a responsible DBA constantly reminding folks that equipment and software are aging, becoming obsolete, impossible to support or fix if broken. Don't commit career suicide by sitting by and not warning fully of the results. Not that you necessarily want to constantly say "I told you so", but be sure you DID tell them so. And don't stake your career by staying too long in this situation. If your company won't listen, get out before the euphemism hits the rotary and save yourself the heartburn.
Rick
Disaster Recovery = Backup ( Backup ( Your Backup ) )
April 14, 2014 at 1:13 am
We're still trying to get everything up to sql server 2008 (not R2) and not currently planning to upgrade from that because the business is up for sale at present.
Most servers are virtual now and the real focus is on upgrading two and a half thousand users from xp and office 2003 to win 7 and office 2010 so sql versions just aren't a priority.
At home I'm just upgrading my xp desktop to win 8.1 - the previous pc was win 98 and both lasted 8 years.
April 14, 2014 at 4:35 am
skeleton567 (4/11/2014)
How long needs to tempered by a responsible DBA constantly reminding folks that equipment and software are aging, becoming obsolete, impossible to support or fix if broken. Don't commit career suicide by sitting by and not warning fully of the results. Not that you necessarily want to constantly say "I told you so", but be sure you DID tell them so. And don't stake your career by staying too long in this situation. If your company won't listen, get out before the euphemism hits the rotary and save yourself the heartburn.
The thing is I think a lot more companies are in this boat than most folks realize. It is only going to get worse and worse as more and more apps come online and now we have had two versions of SQL Server released since Spring of 2012....
April 14, 2014 at 7:14 am
Going from v2000 to v2005 was obviously a leap forward in terms of how SQL Server database are implemented and coded. To a lesser extent v2008 had some compelling features as well. However, v2012 and v2014 are not really game changers from the perspective of how databases are coded or accessed by application developers. These last two versions mostly provide better administration and business intelligence options. I can see why the development and support team for a line of business database would shrug it off.
"Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho
April 14, 2014 at 7:41 am
Eric M Russell (4/14/2014)
Going from v2000 to v2005 was obviously a leap forward in terms of how SQL Server database are implemented and coded. To a lesser extent v2008 had some compelling features as well. However, v2012 and v2014 are not really game changers from the perspective of how databases are coded or accessed by application developers. These last two versions mostly provide better administration and business intelligence options. I can see why the development and support team for a line of business database would shrug it off.
For interal written apps and database tables no. However, the issue becomes for shrink wrap/purchased apps it does become an issue. You have to wait for the software company to certify a new release and service pack. In working with 70 or so plus vendors this is where the bulk of the slowdown to upgrade becomes. We still have one vendor that runs on Win2003 32 bit SQL2005 32 bit. They still haven't certified SQL2008 for all of their apps and they host 20,000 databases in their data center. Of all of the vendors we deal with only 5 have certified SQL2012 so far. THAT is where the first bottleneck is. The second one becomes once they certifiy it typically they will do this for a new release of their product so then you need an IT project and time alloted for installing and testing a new version of their product as well.
April 14, 2014 at 7:42 am
Eric M Russell (4/14/2014)
I can see why the development and support team for a line of business database would shrug it off.
The In-Memory OLTP and In-Memory Columnstore indexes (for data warehouses) are very compelling if you ask me. A lot of people would jump at the chance to extend the life of their existing hardware by adding some memory to the server and upgrading to 2014.
In-Memory OLTP (In-Memory Optimization)
[font="Tahoma"]Bryant E. Byrd, BSSE MCDBA MCAD[/font]
Business Intelligence Administrator
MSBI Administration Blog
April 14, 2014 at 7:44 am
Tatsu (4/14/2014)
Eric M Russell (4/14/2014)
I can see why the development and support team for a line of business database would shrug it off.The In-Memory OLTP and In-Memory Columnstore indexes (for data warehouses) are very compelling if you ask me. A lot of people would jump at the chance to extend the life of their existing hardware by adding some memory to the server and upgrading to 2014.
Yes, but aren't those features only available for ENTERPRISE Edition? If so, not many will be able to justify the huge cost for that version of standard edition. I know we won't be able to.
April 14, 2014 at 11:27 am
I work for a local government agency that has many business functions. Many of the functions don't change that much over the years. We support many databases and applications. We are in the process of moving off of SQL 2000 and 2005. Our technical debt has escalated in the past few years - in part because of the more rapid version pace from Microsoft. At the same time, our budget either is fixed or shrinking. If our platform is not stable for several years, we have will no longer have enough time to handle new customer requirements. Our customers will wait a while, but some will just give up and go elsewhere. In some cases, the "outside" product does not fit well, costs more, and adds to the technical debt. The drive to upgrade is very costly with little or no business benefit. However, the vendor's need to sell new versions is more important than the need to maintain old platforms.
When I was young (before e-readers), I remember that the glue used in book bindings was changed. The new glue was cheeper by a few cents, but would not last very long. Not only was there more profit from selling a book, there was the added profit from selling the book again later because the old book fell apart and was no longer in circulation. It was designed to be thrown away. I stopped collecting books at the time. I never replaced my collection with ebooks in a cloud service - it's just not the same.
It would be nice if there was an effort to provide an application structure that was stable - other than locking into somebody's cloud service. If there was a profit to maintaning a platform, it might be different. Perhaps if we paid maintenace fees instead....
RandyHelpdesk: Perhaps Im not the only one that does not know what you are doing. 😉
April 14, 2014 at 1:06 pm
Markus (4/14/2014)
Yeah, you got me there. I work in the BI space where we don't have a ton of servers. Our DW server, even with Enterprise Edition, is still a far cry cheaper than a Parallel Data Warehouse implementation.
I'll admit, there isn't a lot that's new in 2014 but there are several goodies that weren't publicized as much as as the in-memory features. The new cardinality estimation logic comes to mind but you'll find more here: What's New (Database Engine).
I'm probably coming off as a Microsoft salesperson here 😀 but that's definitely not the case. Since SQL Server 2008 (and maybe even 2005) the upgrade path has been really smooth but only if you're no more than two versions behind. If you get three versions behind then it can be a major headache to upgrade. I like to get on the latest major version as soon as I can get a thorough test plan executed. However, I am in a good position to do that because I don't have third-party applications using my servers and everything I am responsible for is part of the Microsoft BI stack.
[font="Tahoma"]Bryant E. Byrd, BSSE MCDBA MCAD[/font]
Business Intelligence Administrator
MSBI Administration Blog
April 14, 2014 at 1:11 pm
Tatsu (4/14/2014)
Markus (4/14/2014)
[quote-0Yes, but aren't those [In-memory] features only available for ENTERPRISE Edition? If so, not many will be able to justify the huge cost for that version of standard edition. I know we won't be able to.[/quote-0]Yeah, you got me there. I work in the BI space where we don't have a ton of servers. Our DW server, even with Enterprise Edition, is still a far cry cheaper than a Parallel Data Warehouse implementation.
I'll admit, there isn't a lot that's new in 2014 but there are several goodies that weren't publicized as much as as the in-memory features. The new cardinality estimation logic comes to mind but you'll find more here: What's New (Database Engine).
I'm probably coming off as a Microsoft salesperson here 😀 but that's definitely not the case. Since SQL Server 2008 (and maybe even 2005) the upgrade path has been really smooth but only if you're no more than two versions behind. If you get three versions behind then it can be a major headache to upgrade. I like to get on the latest major version as soon as I can get a thorough test plan executed. However, I am in a good position to do that because I don't have third-party applications using my servers and everything I am responsible for is part of the Microsoft BI stack.
I agree with all of your points. I was thrilled when they put backup compression in Standard Edition SQL2008R2. That is one of the big reasons I push folks into an upgrade... LOL. The storage team loves it as well.
April 14, 2014 at 1:43 pm
I wish Microsoft had called this latest release SQL Server 2012 R2. Now those of us not planning to use v2014 in the forseeable future look like fuddy-duddies.
"Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho
April 14, 2014 at 1:48 pm
Eric M Russell (4/14/2014)
I wish Microsoft had called this latest release SQL Server 2012 R2. Now those of us not planning to use v2014 in the forseeable future look like fuddy-duddies.
I am glad they aren't. To me it was too confusing naming it similar. I always make double check with vendors... do you support SQL2008 OR SQL2008R2... Once we had a vendor say yes we support SQL2008R2... when what they heard is SQL2008.... we had a 2008R2 server all built and ready in R2 and when they arrived.. nope... only SQL2008... then... 6 months later the opposite happened... sure we support SQL2008... was going to put their dbs on SQL2008 we already had... they arrrive... and low and behold no they only support SQl2008R2... jeesh.
November 10, 2017 at 2:24 am
GDPR says that all reasonable precautions must be taken to protect personal data taking into account state-of-the-art and cost. A 17 year old DB platform is not state of the art and it's probably running on a similar vintage operating system. Think of the security patches that have occurred since that combination went out of extended support.
Even if that system does not contain personal information it is a potential weak point that could be used as a backdoor into other systems. €20million or 4% of organisation revenue should provide the leverage to upgrade or deprecate
November 10, 2017 at 2:24 am
I think this is an even more important question now that we're entering what, bi-yearly or is it now yearly release cycles? Does anyone know how the CAL licencing works these days (we use SQL Express, currently 2016, so doesn't apply - small company, small databases - but may at some point)? If we get SQL 2017 Standard with an additional 10 CALs (Core licensing is way out of our price range) can these CALs be moved to SQL 2018/19 when upgraded, or is that a new set of CALs and 2017's are essentially useless?
Not that we'd go with such an upgrade path. To answer the original question I would expect at least 5 years service out of a version. Personally I look at each new release and see what it offers. We'd still be on 2012 if not for the native JSON parsing that 2016 SP1 introduced.
November 10, 2017 at 3:06 am
Newbie poster here but long time voyeur!
What are people’s experiences in jumping over versions when upgrading? We are currently looking at going from 2008R2 to 2016 for a number of applications. As far as I can see the 2016 Migration Assistant (upgrade advisor!) doesn't even consider 2008R2. Would you upgrade to 2012/2014 and then do a separate upgrade to 2016 or just go straight to 2016?
Viewing 15 posts - 31 through 45 (of 57 total)
You must be logged in to reply to this topic. Login to reply