July 5, 2014 at 11:27 am
Comments posted to this topic are about the item Cloud Skills
July 6, 2014 at 11:52 pm
I think SQL Server as a Box will stay around for a long time, -even though we do Azure. OLTP Load as an example. Limitations as you mention.
As the SQL Server Version's has changed over the year's, there has been new areas to learn, -going from cocommandlinend Query Analyzer to management Studio, -get Dynamics management views. Sure there has been a lot to learn with every New Version.
For the current Version, Hyper-V, SMB 3 and Storage Spaces, In Memory solution's, "hybrid cloud", tuning on Hyper-V for SQL Server (I run around50 always on installations on Hyper-V With Virtual drives.., ehm, files:))
New Security models and .........., I always been thinking that beeing a DBA involves hardware and Storage, -understand NUMA and SPLAT, Volume alignment, lun.. And not to forget Clustering. Active Directory and certificates.
But if we look at the DBA role as you describe, sure we will loose in the long run
July 7, 2014 at 12:02 am
I find it very hard to see how Microsoft will be adding value to someone generating thousands of virtual machines until it comes up with a way to license its software at a reasonable cost per unit.
They have figured out this puzzle before.
July 7, 2014 at 1:35 am
I was just going to post a question in the forum about this topic exactly.
And not related to the db engine only, but even to the analytics.
I have no clear thoughts to share here, at the moment, but my feeling is that that the db engine part is moving toward a SMB market (or, better, the SMB market is going to become its main adopter), the cloud is being addressed to large distributed businesses, and the analytics are going left in favour of competitors.
July 7, 2014 at 2:44 am
Those whose primary expertise is in SQL Server appear to have the benefits and disadvantages of working in a single system. I know that this is a simplistic view but it does mean that the knowledge required is more clearly defined than some others and also it does tie them to a single product in the short term.
I am comparing this to application development where there are different application types; services, web services, web applications, desktop applications etc. With this comes the difficulties of a far wider range of possible technologies that means that there is less tie to the success or failure of a single product but, I would argue, more technologies to employ.
In some respects this can explain the depth of knowledge that DBAs can have whereas software developers often need a breadth of knowledge. Neither job or career is easier. Just different challenges.
Gaz
-- Stop your grinnin' and drop your linen...they're everywhere!!!
July 7, 2014 at 3:27 am
My impression is that Microsoft does have internal divisions that fight each other for supremacy and at the moment the cloud is in the ascendancy. I think the vision within the cloud section is that they are trying to reduce the need for configuration experts. I would suggest this however will take a significant time to achieve.
That said I think there is a rise in Dev Opps and most of the members here are ideal candidates for such a role. So while in the long term the role of the DBA as it is today will alter probably getting less involved the need for intelligent mathematical individuals to connect with and use databases is only going to increase.
This is a change that is going to touch all professions not just DBAs.
cloudydatablog.net
July 7, 2014 at 3:47 am
One worrisome scenario is SME's or even larger companies going to the cloud "because everyone else is". I remember a conversation back in 2000 where one client asked the web/ISP company I was working for to transform their entire website to XML, as they'd heard it was the next version of HTML. 😀
Good article, creating lots of discussion in our DBA team this morning.
qh
July 7, 2014 at 3:54 am
quackhandle1975 (7/7/2014)
One worrisome scenario is SME's or even larger companies going to the cloud "because everyone else is". I remember a conversation back in 2000 where one client asked the web/ISP company I was working for to transform their entire website to XML, as they'd heard it was the next version of HTML. 😀...
Ah, but that's where our responsibility as professionals comes in to play!!!
Gaz
-- Stop your grinnin' and drop your linen...they're everywhere!!!
July 7, 2014 at 6:55 am
The SME thing is interesting, it certainly seems to look like that pattern of moving to the cloud is there for the SME's that work at MS.
July 7, 2014 at 6:59 am
Comparing SQL Server to SQL Azure is like comparing traditional C++/C# forms to ASP.NET. There are some applications where moving to SQL Azure is an evolutionary leap, but there still be plenty of applications where sticking with SQL Server is the best choice, not only for existing implementations but also new developmet going forward.
"Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho
July 7, 2014 at 7:08 am
Eric M Russell (7/7/2014)
Comparing SQL Server to SQL Azure is like comparing traditional C++/C# forms to ASP.NET. There are some applications where moving to SQL Azure is an evolutionary leap, but there still be plenty of applications where sticking with SQL Server is the best choice, not only for existing implementations but also new developmet going forward.
I agree.
One example might be an application on a mobile device (laptop or smaller footprint) where data is captured locally and synchronised when a signal is available. I have done this and it is a key requirement when engineers are working where they might not be a connection always available. In this case then a local (small) database is often ideal. I don't think that McLaren racing team would put up with zero data if their connection went down.
Another example might be those databases that are used in significant data analysis. Although, I have not looked into the facilities for doing this using Azure and I may well be proven wrong. If only eventually.
Gaz
-- Stop your grinnin' and drop your linen...they're everywhere!!!
July 7, 2014 at 7:10 am
Honestly when you look at Microsoft's larger play it really does feel like they are attempting to reduce the number of DBA's and DB Developers.
Their Entity Framework changes and marketing is all about letting the middle/front end developers drive the data model using a code first decision. I understand MS needs to compete against other ORM solutions but we all know how "well" ORM creates T-SQL code.
Then when you look at their push to move databases to Azure, again I feel like the overall pattern is to reduce the number of database experts and push a lot towards developers.
Totally fine, MS is a company and needs to do what is in their best interests. And I understand there are some situations where security,etc will prevent cloud usage.
To me though it feels like MS is attempting to push more into the developers hands.
July 7, 2014 at 7:14 am
Chord, if the ORM's get better, is that better or worse for a DBA - ugly question isn't it? Businesses get pushed into hiring a lot of silo'd experts (DBA's being one of them), so I imagine there is always interest in reducing that cost, but that never feels like the reason for the cloud. Part is that grand possibility of scale out, part is dumbing down data access so they can move faster, and I think a big part is just moving to the pay by the month model to avoid up front costs for hardware/licenses.
July 7, 2014 at 7:23 am
Andy Warren (7/7/2014)
Chord, if the ORM's get better, is that better or worse for a DBA - ugly question isn't it?
Oh I agree it's an ugly question. It will be worse for DBA's.
With Database as a Service and ORM's (if both keep maturing as expected), it honestly paints an ugly picture for a traditional DBA.
July 7, 2014 at 7:33 am
Chord77 (7/7/2014)
Andy Warren (7/7/2014)
Chord, if the ORM's get better, is that better or worse for a DBA - ugly question isn't it?Oh I agree it's an ugly question. It will be worse for DBA's.
With Database as a Service and ORM's (if both keep maturing as expected), it honestly paints an ugly picture for a traditional DBA.
Not sure that I agree. I think that a lot of the smaller databases with low performance constraints will drop of your DBA radars leaving the "harder" work to DBAs. I think that this is akin to progressing to assembler, 3GLs, 4GLs, etc. There is so much to be done that any kind of abstraction frees up experts to utilise their expertise as opposed to do hum drum work that can be either automated or left, in this case, to developers.
The same can be true of the memory management in .NET. Most applications, services etc. do not need to concern themselves with memory and resource usage as the default garbage collection and process recycling is good enough for most applications and so forth. You will find that there are few application developers today that really can do memory and resource management. And as for multi-threading, less people can do it well enough than think that they can (cleaning up after so many "experts" has taught me that one).
Gaz
-- Stop your grinnin' and drop your linen...they're everywhere!!!
Viewing 15 posts - 1 through 15 (of 27 total)
You must be logged in to reply to this topic. Login to reply