March 26, 2018 at 3:44 am
I have total 5 year exp In Sql, ssrs, ssis, ssas(3 year).
I was thinking what to fo for next.
Learning new reporting tools(currently looking into it)
Deep learning of data modeling and dataware house concepts?.
Big data ?
please give suggestions.
If any field I can try into.. .
March 26, 2018 at 5:01 am
Master one of the aspects you already know, is another option
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
March 26, 2018 at 6:32 am
I would suggest learning some of the trending technologies used with those buzz words. Python is the most easiest that comes to mind. Another may be diving into the various data stores. You could start with Azure Data Lake Store being you are mostly working on the Microsoft stack. The chances of you running into that as a Microsoft SQL person is highly possible in the future as more organizations shift to the cloud from on-prem. Then there is the other visualization tools that are growing in popularity over SSRS like Tableau and PowerBI that you can start getting into as well.
Course, Gila's option is good too. The only argument I would provide to that is mastering one of those existing technologies also may mean less time for a new technology. Both have their pro's and con's. I would opt for the new stuff myself because tech is changing so fast that I rarely run into positions looking for masters of a single piece of tech versus helping shift a company to a new piece of tech their masters refuse to learn or adapt. 😉
March 26, 2018 at 7:39 am
super48 - Monday, March 26, 2018 3:44 AMI have total 5 year exp In Sql, ssrs, ssis, ssas(3 year).
I was thinking what to fo for next.
8 Months ago, I suggested that "BI" would help put you in demand because it's always a hot topic. What have you actually done in that area? For example, have you been able to do predictive analysis?
Have you been able to use SQL (T-SQL, actually) to augment your skills in SSRS, SSIS, SSAS? Do you know how to do things such as Dynamic Cross Tabs? Can you use a Tally or Numbers table in your sleep to encode or decode things like Luhn Mod 10 Checksums and UccEan checksums. Do you know how to resolve the problem of overlapping dates to come up with a single start and end date to represent the total range? Do you know how to properly solve "Catch All" queries without introducing possibilities for SQL Injection? Do you know why you might want to make another index that has the identical keys as the Clustered Index? Do you know how to make, use, and maintain Partitioned Tables and Partitioned Views and why you might want to use one over the other? Do you think any form of partitioning should be done for performance of code Do you know the difference between an mTVF and an iTVF and why you should prefer one over the other? Do you know how to build an inline Scalar Function and why you would want to do so? Do you know how to import a horizontally based spreadsheet with an unknown number of columns and column names and normalize the data in a table? Do you know how to do running totals and "data smears" to fill in missing data from an import? In the absence of production data, do you know how to make a million rows of random but constrained test data without using a While Loop or Recursive CTE and why you want to avoid both? Do you know how to do a "Last Name" search on a "Full Name" column and still use a SARGable predicate? Do you know what a "blocking operation" in a cascading CTE is and why you'd want to use such a thing especially when "Divide'n'Conquer" methods for pre-aggregation? Have you read Grant Fritchey's book on execution plans or any of Itzik Ben-Gan's books? And do all of that with nasty fast performance in mind? (And all that's just scratching the surface).
The bottom line question would be... after 5 years of experience in T-SQL, what do you actually know to do with it?
And, no... I'm NOT making fun of you for any reason. If I did, I'd be making fun of myself because, at the 5 year mark in my career, I had very successfully written a World Wide Call Accounting system that imported data from multiple disparate sources across the world and exported to more and it was all done in T-SQL I was thinking that I actually knew T-SQL after that and I was dead wrong. All of that had only qualified me to learn more about the tool and, after 23 years of using the tool, I still learn something new almost every day.
There are a whole lot of bright shiny tools out there that you could learn and many will come and go just like most of the older ones have (certain tools, such as Python, are likely going to stay around for a very, very, long time). The one thing that hasn't changed is the need to store and access data in an accurate, secure, and high performance manner. It's not going to go away.
So my suggestion for the implied "What's Next" question is similar to what Gail suggested above except that I'll also suggest the tool to do a much deeper dive on. Learn T-SQL and all that's related to it. Understand (and I certainly didn't at my 5 year mark) that there's a whole lot to know that you may not even know about right now. SSIS, SSRS, and SSAS and all tools like them can all seriously be benefitted by a deep knowledge of T-SQL and the tools available in it and some of the tricks of the trade.
--Jeff Moden
Change is inevitable... Change for the better is not.
March 26, 2018 at 9:35 am
xsevensinzx - Monday, March 26, 2018 6:32 AMI would opt for the new stuff myself because tech is changing so fast that I rarely run into positions looking for masters of a single piece of tech versus helping shift a company to a new piece of tech their masters refuse to learn or adapt. 😉
I suggest 'master', because in most cases 5 years of using 4 different techs will get someone to roughly intermediate in them (assuming no extra studying). I'd rather hire someone who knows T-SQL well over someone who knows a bit about T-SQL and Python and reporting and a couple other things.
A 'master' who refuses to learn new stuff is either in denial or being stupid. Mastering an area doesn't mean you can't or shouldn't learn others.
Jeff Moden - Monday, March 26, 2018 7:39 AM8 Months ago, I suggested that "BI" would help put you in demand because it's always a hot topic. What have you actually done in that area? For example, have you been able to do predictive analysis?
Predictive analytics isn't BI 🙂 (not by most definitions of the term)
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
March 26, 2018 at 10:19 am
GilaMonster - Monday, March 26, 2018 9:35 AMxsevensinzx - Monday, March 26, 2018 6:32 AMI would opt for the new stuff myself because tech is changing so fast that I rarely run into positions looking for masters of a single piece of tech versus helping shift a company to a new piece of tech their masters refuse to learn or adapt. 😉I suggest 'master', because in most cases 5 years of using 4 different techs will get someone to roughly intermediate in them (assuming no extra studying). I'd rather hire someone who knows T-SQL well over someone who knows a bit about T-SQL and Python and reporting and a couple other things.
Sure, but that's you. In my experience, those that are dabbling in multiple technologies are looking for people who can go across multiple technologies too. That does not mean no one is looking for that TSQL expert or rather hire someone with better knowledge in TSQL versus some in TSQL and the other in Python. But the days of just finding a gig to only do TSQL 24-7 is winding down a lot. I don't think I've talked to an organization in the past 6 months that is hiring just a SQL Server guy anymore or just a TSQL guy anymore. But then again, I'm more focused on the BI/Analytical side of things too, which is expanding a great deal in new tech right now.
On the BI stuff. I think that depends on the organization. In the same sense, many organizations view BI as analytics among other things simply because any analysis on the business data is technically still business intelligence.
March 26, 2018 at 3:22 pm
GilaMonster - Monday, March 26, 2018 9:35 AMxsevensinzx - Monday, March 26, 2018 6:32 AMI would opt for the new stuff myself because tech is changing so fast that I rarely run into positions looking for masters of a single piece of tech versus helping shift a company to a new piece of tech their masters refuse to learn or adapt. 😉I suggest 'master', because in most cases 5 years of using 4 different techs will get someone to roughly intermediate in them (assuming no extra studying). I'd rather hire someone who knows T-SQL well over someone who knows a bit about T-SQL and Python and reporting and a couple other things.
A 'master' who refuses to learn new stuff is either in denial or being stupid. Mastering an area doesn't mean you can't or shouldn't learn others.
Jeff Moden - Monday, March 26, 2018 7:39 AM8 Months ago, I suggested that "BI" would help put you in demand because it's always a hot topic. What have you actually done in that area? For example, have you been able to do predictive analysis?Predictive analytics isn't BI 🙂 (not by most definitions of the term)
Apparently were from different schools on that subject.
--Jeff Moden
Change is inevitable... Change for the better is not.
March 26, 2018 at 9:26 pm
xsevensinzx - Monday, March 26, 2018 10:19 AMGilaMonster - Monday, March 26, 2018 9:35 AMxsevensinzx - Monday, March 26, 2018 6:32 AMI would opt for the new stuff myself because tech is changing so fast that I rarely run into positions looking for masters of a single piece of tech versus helping shift a company to a new piece of tech their masters refuse to learn or adapt. 😉I suggest 'master', because in most cases 5 years of using 4 different techs will get someone to roughly intermediate in them (assuming no extra studying). I'd rather hire someone who knows T-SQL well over someone who knows a bit about T-SQL and Python and reporting and a couple other things.
Sure, but that's you. In my experience, those that are dabbling in multiple technologies are looking for people who can go across multiple technologies too. That does not mean no one is looking for that TSQL expert or rather hire someone with better knowledge in TSQL versus some in TSQL and the other in Python. But the days of just finding a gig to only do TSQL 24-7 is winding down a lot. I don't think I've talked to an organization in the past 6 months that is hiring just a SQL Server guy anymore or just a TSQL guy anymore. But then again, I'm more focused on the BI/Analytical side of things too, which is expanding a great deal in new tech right now.
On the BI stuff. I think that depends on the organization. In the same sense, many organizations view BI as analytics among other things simply because any analysis on the business data is technically still business intelligence.
Yeah happening job openings looking for guys with multiple skills.
You know that this if you ssrs they will ask you know power bi?.
Though sql is mandatory.
March 27, 2018 at 8:03 am
Predictive analytics isn't BI 🙂 (not by most definitions of the term)
Apparently were from different schools on that subject.
I agree with Gail on this. I'll conceded at the outset that there is no agreed definition of BI. But most definitions talk about organizing your company's information to make better decisions. Predicative analytics can be a part of that, but it isn't often included, because it's a subset. I have designed many BI solutions, but none have had predictive analytics. It is an area I'm trying to learn more about as I have seen some successful uses of this kind of data, though not in a MS BI solution.
March 27, 2018 at 9:09 am
RonKyle - Tuesday, March 27, 2018 8:03 AMPredictive analytics isn't BI 🙂 (not by most definitions of the term)
Apparently were from different schools on that subject.
I agree with Gail on this. I'll conceded at the outset that there is no agreed definition of BI. But most definitions talk about organizing your company's information to make better decisions. Predicative analytics can be a part of that, but it isn't often included, because it's a subset. I have designed many BI solutions, but none have had predictive analytics. It is an area I'm trying to learn more about as I have seen some successful uses of this kind of data, though not in a MS BI solution.
Heh.... you do have to organize the data and then you do have to analyze it to make any decisions... better or not. To me, predicative analytics isn't a subset for BI... it's the whole reason for BI. Without it, BI becomes an oxymoron and a waste of disk space, IMHO.
But, disagreement or not, I do agree that if the data isn't organized, you can't do a thing with it.
--Jeff Moden
Change is inevitable... Change for the better is not.
March 27, 2018 at 9:22 am
We may differ on what predictive analytics is. BI can be used to answer such questions as who are my biggest customers or best technicians by evaluation rating. PA answers questions such as what is the likelihood of a machine part failing based on these readings. Some of this could be addressed by the data mining component of SSAS. In the absence of that, I think a case that BI=PA is a tough one to make.
Bringing this back to the original poster, as you can see, there is still a lot to learn. I have great skills in TSQL, Database design, SSIS, and SSAS, and MDX. I have some experience with SSRS and Power BI. I'm not bad at the database maintenance aspects, but I read Gail Shaw's and others' posts with great interest as I know they know much more than I. But even after all these years, I'm hesitant with the with the word export, because I know how much I don't know. For example, in SSIS, I haven't had to do much scripting, certainly nothing compared to the days in which I worked with DTS. But I also haven't seen a good source to teach me what is possible and realistic that can't be done some other way.
March 28, 2018 at 7:31 am
RonKyle - Tuesday, March 27, 2018 9:22 AMWe may differ on what predictive analytics is. BI can be used to answer such questions as who are my biggest customers or best technicians by evaluation rating. PA answers questions such as what is the likelihood of a machine part failing based on these readings. Some of this could be addressed by the data mining component of SSAS. In the absence of that, I think a case that BI=PA is a tough one to make.
Well, using the term predictive analytics is a bit too specific. Machine learning is surely a segmentation, but still within business intelligence. Predictive analytics is just one subset of what you can do with machine learning in terms of the supervised learning field. There is the whole unsupervised learning aspect too. But speaking on the supervised learning more, you do have classification that is apart of that. Who is big or not in your example is a form of classification. The same for ranking based on some attributes or metrics.
In the unsupervised learning route, you have clustering to further find those inherent groupings that can lead to further classification among other associated methods to understand if people are doing X, they are also doing Y with those big customers. Which might I add, would only further enhance your example because you're likely taking a simple approach to define big based on the highest metric, versus using other metrics (i.e.: weights) to determine if that customer truly is big or small.
All of this is still within BI because it's all taking approaches to extract intel from the data, just with different methodologies. Hence why SSAS has the whole machine learning component to complete the Microsoft BI stack. In that sense, data science is just another classification of BI with a specific focus on machine learning as opposed to just using the standard functions to aggregate/summarize your data for canned reporting.
March 28, 2018 at 7:35 am
All of this is still within BI. In that sense, data science is just another classification of BI with a specific focus on machine learning as opposed to just using the standard functions to aggregate/summarize your data for canned reporting.
Agreed with the entire direction of your comment. I should have followed up BI=!PA with PA subset BI.
March 28, 2018 at 7:42 am
I will say that the whole AI aspect and how it works with machine learning to execute is the only wonky part of this whole debate. I don't think that's BI in the sense, but then again, that's only because you replace the human with a machine on the decision making process on top of a greater focus on automation of those insights and decisions to boot.
For example, it's machine learning gathering historic data on a user, then AI using that historic data to determine a recommendation, then machine learning executing that recommendation in an automated fashion. Dono if that really falls in the realm of BI as much as it does in something else.
Viewing 14 posts - 1 through 13 (of 13 total)
You must be logged in to reply to this topic. Login to reply