October 27, 2015 at 5:31 am
tguerrerovl (10/26/2015)
I really enjoyed this article... a good understanding of what we are talking about:😛
Ahh, good stuff.
October 27, 2015 at 7:43 am
Gary Varga (10/27/2015)
I pay little attention to these lists as they often highlight niche skills that are becoming more commonplace or being rolled into other roles over time.
When an industry publication mentions a job as being the "hot" thing next year, what that really means is that hiring managers now have to sift through 10x the number of resumes just to find that one single qualified candidate.
"Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho
October 27, 2015 at 8:20 am
xsevensinzx (10/26/2015)
Eric M Russell (10/26/2015)
xsevensinzx (10/26/2015)
Eric M Russell (10/26/2015)
..If you look through the jobs, many are development positions, but there are some good data ones. Data modeler, BI Analyst, Data Architect, Big Data Engineer, and, of course, Data Scientist, are all listed.
..
I notice that DBA isn't listed, though I don't think DBAs are going to disappear anytime soon. However I think the traditional DBA that manages a database like a sysadmin is a job that is morphing. I suspect that DBAs will be asked to do more, either in development or analysis, or they'll be paid less in the future..
Unless an organization is large and diverse, or perhaps a smaller organization that specializes in data analytics (ie: client based consulting), I doubt many of these pure "Big Data" and "Data Scientist" positions are built to last. For the vast majority of organizations, hosting your own dedicated in-house Big Data Developer or Data Scientist makes about as much sense as hosting your own in-house cloud data center. For this to be a true full-time permanent position, I suspect that 80% of the day to day responsibilities would be traditional DBA tasks.
What do you mean? Data Scientist positions are typically involving a lot of complex areas that's outside the DBA's scope. All the DBA can do is ingest, provide and secure the data and the infrastructure. In terms of developing and running complex algorithms on the data, there is not much a DBA can do in those areas nor evolve them in advance statistical models nor maintain them over time.
As databases move to the cloud, database administrators will branch out into other related areas, like developing and running complex algorithms. Anyone with a background in database technology, a grasp of 200 level basic statistics, sufficient time on their hands, and a subscription to Safari Books Online can do "data science" on a practical commercial level.
I agree. Anyone with a background in data can leverage their knowledge into other data related fields. But, let's be honest here. Statistics, probability, predictive analytics, regression modeling and so forth are not something you just pick up over night nor an area where someone is going to trust their 1 million dollar budget in your hands to forecast for the next financial quarter.
It takes a real set of qualifications and skills to fill those roles and have someone in there that understands what they are doing. If you wouldn't hire a DBA with 10 years of experience who really didn't know much about SQL Server, why in the world would you hire someone to do the same with 0 years of experience on analyst side of things simply because they can manage a RDBMS? Especially if what you're doing with data science involves real money? You really can't compare DBA's to Data Scientist in that way. Two totally different disciplines with data.
Sorry if I sound a bit sensitive on the topic. Sounded like you're saying that anyone can do it and it's a irrelevant field. I work with a bunch of data scientist who are very much on another level with the data that I have yet to see any DBA touch. 😛
I would say that entirely depends on the DBA and what experience they have. Often time DBA's are asked to do a lot more than just keep the servers running.
October 27, 2015 at 8:42 am
My (admittedly un-scientific) belief is that only a small percentage of organizations in private (or public) sector engage in what could be genuinely called "data science". Whatever predictive modeling they do, can be done by a 3rd party product that the DBA with intermediate level analytical skills can install, configure, and maintain. SSIS itself comes out of the box with dataflow tasks for building data mining models. A genuine data scientist would be bored (and underaid) working full time in a staff position for a county government or medium sized retailer.
This goes back to my analogy of the "nutritionist" working for McDonalds.
McDonald's Nutritionist Cindy Goody: Our Food Is Healthy
http://www.huffingtonpost.com/2013/03/29/mcdonalds-nutrition_n_2980991.html
"Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho
October 27, 2015 at 9:14 am
Eric M Russell (10/27/2015)
My (admittedly un-scientific) belief is that only a small percentage of organizations in private (or public) sector engage in what could be genuinely called "data science". Whatever predictive modeling they do, can be done by a 3rd party product that the DBA with intermediate level analytical skills can install, configure, and maintain. SSIS itself comes out of the box with dataflow tasks for building data mining models. A genuine data scientist would be bored (and underaid) working full time in a staff position for a county government or medium sized retailer.This goes back to my analogy of the "nutritionist" working for McDonalds.
McDonald's Nutritionist Cindy Goody: Our Food Is Healthy
http://www.huffingtonpost.com/2013/03/29/mcdonalds-nutrition_n_2980991.html
Yeah, data science or even advanced analytics is certainly a niche area for a lot of companies. Many do rely on third-party BI solutions that help them understand how they are succeeding or failing. But, this area of focus is on the rise now that data is becoming more and more accessible. Anyone can get up in running with a database these days, and there are plenty of solutions to do BI with that data. So, the need for BI experts is rising too.
That being said, yes, there are third-party predictive analytic tools that make things easy and automate. The problem that a lot of organizations that employ data scientist for is that very few can make sense on the results they are getting and mostly importantly that most DBA's can't answer is why.
Unfortunately, not all models can fit and work with all data. Everything is not a nail for this hammer.
For example, anyone can leverage SSAS data mining to create a simple view of the data with their plug-in-play algorithms to do forecasting. I can do this within a hour and have a nice little time series graph that is predicting the sales for the coming year. I can explain the results because they are as clear as they are on the sexy graph I have. However, is you ask me how and why? I have no idea why because it's just what SSAS gave me.
Fitting this models to the data and truly being able to interpret the results is what's lacking from the data professionals who are not as specialized in statistics and so forth is what's dangerous. Why is it dangerous? Because you have absolutely no idea why you are getting the results you are getting. You're depending on the machine (e.g.: SSAS) to do it all for you.
Can you dive deeper into SSAS to better understand the time series algorithm? Yes. There is documentation and if you have C++ chops, you can take it to the next level too. But, you still have to know what you're looking at when you dive into this ocean.
So, data scientist and other related fields are called into action to make sense of it all and take it to the next level. Unfortunately, tools like SSAS are not commonly used in those areas due to the lack of flexibility and the high costs of the Microsoft stack. That's why R, Python, SAS and others need data scientist to leverage these tools and develop analytical solutions that can be interpreted.
October 27, 2015 at 9:57 am
xsevensinzx (10/26/2015)
Jeff Moden (10/26/2015)
xsevensinzx (10/26/2015)
Eric M Russell (10/26/2015)
..If you look through the jobs, many are development positions, but there are some good data ones. Data modeler, BI Analyst, Data Architect, Big Data Engineer, and, of course, Data Scientist, are all listed.
..
I notice that DBA isn't listed, though I don't think DBAs are going to disappear anytime soon. However I think the traditional DBA that manages a database like a sysadmin is a job that is morphing. I suspect that DBAs will be asked to do more, either in development or analysis, or they'll be paid less in the future..
Unless an organization is large and diverse, or perhaps a smaller organization that specializes in data analytics (ie: client based consulting), I doubt many of these pure "Big Data" and "Data Scientist" positions are built to last. For the vast majority of organizations, hosting your own dedicated in-house Big Data Developer or Data Scientist makes about as much sense as hosting your own in-house cloud data center. For this to be a true full-time permanent position, I suspect that 80% of the day to day responsibilities would be traditional DBA tasks.
What do you mean? Data Scientist positions are typically involving a lot of complex areas that's outside the DBA's scope. All the DBA can do is ingest, provide and secure the data and the infrastructure. In terms of developing and running complex algorithms on the data, there is not much a DBA can do in those areas nor evolve them in advance statistical models nor maintain them over time.
Being a DBA that regularly comes up with such complex algorithms on a regular basis and can explain such complexities so that even previously unaware neophytes can actually understand them, you'll have to understand when I say I take strong exception to that. 😉
Hah, guess I walked into that one. I'm sure you have written linear regression in SQL and then some Jeff. :w00t:
But, what I'm trying to say here is that you can't possibly just toss data science and statistics aside as if there is no true discipline there. Like as if because you have a DBA then all your statistical modeling, predictive analytics and forecasting problems are solved. The same goes for the other end. Working in data science does not mean all your DBA problems are solved either.
Now, I agree with both of those notions. According to most of the folks that I've interviewed lately, you might not have solved all your DBA problems even by hiring a "DBA".:sick:
--Jeff Moden
October 27, 2015 at 6:37 pm
The hot jobs comes from the fact there is demand. This has nothing to do with quality, with whether this is a good idea, but more that these are positions companies are looking for, but also looking to pay more.
Not every company, not every job. Plenty of companies might not hire these positions or pay more.
However there is opportunity here for some. Whether you take advantage of it, or do a good job if you get the opportunity is separate from the opportunities available.
Follow me on Twitter: http://www.twitter.com/way0utwest
Forum Etiquette: How to post data/code on a forum to get the best help
My Blog: www.voiceofthedba.com
November 4, 2015 at 3:31 pm
Spent all last week at DevIntersection 2015, concentrating on the SQL track since I am a DBA. I can tell you based on the questions coming from other attendees that I am not worried about my career tanking any time soon. DBA - it's not sexy, but try living a week without one. I guess we're kind of like a liver in that regard. Maybe a kidney? You get the point.
And then there is good 'ol LinkedIn, where at least 10% of my contacts are local recruiters. I don't know about your city, but mine...they need DBAs as evidenced by the number of interview offers I have to turn down every month. The DBA career track may not be a fast growing, super high paying, top of anyone's list job. But it's my job and it's here to stay.
November 4, 2015 at 3:39 pm
DBA isn't one of the 10 hot ones, but it's not that far behind.
Follow me on Twitter: http://www.twitter.com/way0utwest
Forum Etiquette: How to post data/code on a forum to get the best help
My Blog: www.voiceofthedba.com
November 4, 2015 at 8:25 pm
sgtmango333 (11/4/2015)
Spent all last week at DevIntersection 2015, concentrating on the SQL track since I am a DBA. I can tell you based on the questions coming from other attendees that I am not worried about my career tanking any time soon. DBA - it's not sexy, but try living a week without one. I guess we're kind of like a liver in that regard. Maybe a kidney? You get the point.And then there is good 'ol LinkedIn, where at least 10% of my contacts are local recruiters. I don't know about your city, but mine...they need DBAs as evidenced by the number of interview offers I have to turn down every month. The DBA career track may not be a fast growing, super high paying, top of anyone's list job. But it's my job and it's here to stay.
What were some of the questions? Would be interesting to hear more.
Viewing 10 posts - 16 through 24 (of 24 total)
You must be logged in to reply to this topic. Login to reply