Data Wareshousing

Zen of data modelling in Hadoop

Zen of data modelling in Hadoop

The Zen of Python is well known tongue-in-cheek guidelines to writing Python code. If you haven’t read it, I would highly recommend reading it here Zen of Python.

There’s an “ester egg” in Python REPL. Type the following import into Python REPL and it will printout the Zen of Python.

import this

Beautiful is better than ugly.
Explicit is better than implicit.
Simple is better than complex.
Complex is better than complicated.
Flat is better than nested.
Sparse is better than dense.
Readability counts.
Special cases aren’t special enough to break the rules.
Although practicality beats purity.
Errors should never pass silently.
Unless explicitly silenced.
In the face of ambiguity, refuse the temptation to guess.
There should be one-and preferably only one –obvious way to do it.
Although that way may not be obvious at first unless you’re Dutch.
Now is better than never.
Although never is often better than right now.
If the implementation is hard to explain, it’s a bad idea.
If the implementation is easy to explain, it may be a good idea.
Namespaces are one honking great idea -let’s do more of those!

I have been pondering about how would you model your data in Hadoop. There are no clear guideline or hard and fast rules. As in, if it is an OLTP database use 3NF or if you are building warehouse, use Kimball methodology. I had the same dilemma when SSAS Tabular came around in SQL Server 2012. The conclusion that I reached was that it was somewhere between the two extremes: a fully normalized > = 3NF and flat-wide-deep denormalized mode.
Hadoop ecosystem also throws additional complexities in the mix – what file format to choose (Avro, Paraquet, ORC, Sequence and so on), what compression works best (Snappy, bzip etc), access patterns (is it scan, is it more like search), redundancy, data distribution,partitioning, distributed/parallel processing and so on. Some would give the golden answer “it depends”. However, I think there are some basic guidelines we can follow (while, at the same time, not disagreeing with the fact that it depends on the use case).

So in the same vein as  Zen of Python I would like to present my take on Zen of data modelling in Hadoop. In future posts – hopefully – I will expand further on each one.
So here it is

Zen of Data Modelling in Hadoop

Scans are better than joins
Distributed is better than centralised
Tabular view is better than nested
Redundancy is better than lookups
Compact is better than sparse
End user queries matter
No model is appropriate for all use cases so schema changes be graceful
Multiple views of the data are okay
Access to all data is not always needed, filtering at the source is great
Processing is better done closer to data
Chunk data into large blocks
ETL is not the end, so don’t get hung up on it, ETL  exists to serve data
Good enough is better than none at all, Better is acceptable than Best

So there you have it. As mentioned previously, the model will depend on the situation and lets hope these will be helpful starting point. They come with full disclaimer – Use at your own risk and are always subject to change 🙂

Changing role of IT in the BI world

BI is Dead, Long live BI

Timo Elliot (blog|twitter) recently published a blog BI is Dead which draws from Gartner’s Magic Quadrant report and a detailed report.

The main take away from the post (which includes references from the Gartner research ,so not all are Timo’s points) include:

  •  The self service analytics tools have matured enough that the involvement of IT in BI and analytics projects is deemed optional. IT don’t need to model the data upfront (which needed gathering all or partial analytics requirements to start with), analysts can prototype and test it themselves.
  •  The balance of power for BI and analytics platform buying decisions has gradually shifted from IT to business.

  • BI and analytics tools which do require intervention from IT are not considered BI anymore; they are Enterprise Reporting Based Platforms. Admittedly, they take most share of the BI market. In other words Gartner has updated the definition of BI.

  • The bold headline – “BI is Dead” – is to gain attention on the changing landscape of BI tools.

  •  Organizations who do not embrace the new definition of BI, run the danger of turning into BI-nosaurs.

  •  Having single view of data through an EDW is pointless/extremely hard. This and this from Curt Monash also support this line of thinking.

  •  Most organizations believe that IT has role to play in BI although majority want the responsibility of authoring the content to end users. This has always been the holy grail of BI.

So it is really dead?

Well-not necessarily. It’s dead in the sense that PCs are dead as we know them from 1980 or before. PCs have clearly evolved: they don’t look like how they used to look and many aspects of PC have been democratized. Trivial things which needed programmer in those days can now be done by users themselves. Heck, you can even upgrade the hardware if you are into that sort of thing.

I think that’s exactly what is happening in the BI and analytics world. The self service analytics tools have evolved to the point that many of the trivial data munging tasks can be done by the analysts themselves and what they can do with these new tools does not look at all like how they have been doing it. Does that mean the BI is dead? No- the fundamental analytics is still the same. It has evolved just like the PCs and the role of IT has changed.

Role of IT in the new BI world

The changing role of IT in BI and analytics can be best described from picture below.

Changing role of IT in the BI world

IT is now viewed as (and rightly so) as data facilitators. They make data available to the analysts in palatable form. The responsibility of modelling it,authoring presentation components and analysis lies with analysts.

Does that mean the job of IT got easier?

Most-Definitely-Not, on the contrary it has got even more complicated. With the influx of new tools, the demand for data has dramatically increased. The analysts want to analyse all sorts of data, from all sorts of unlikely sources and with all sorts of analytical methods. Their expectations from IT is to facilitate this process which is where ITs job has got a lot harder. We have to deal with ever increasing data sizes, from ever expanding sources, and support ever changing analytical tools.

What else is keeping BI alive?

If we assume the following definition of BI,then here are more reasons why BI is definitely not dead in its current incarnation.

BI
:
an umbrella term to describe “concepts and methods to improve business decision making by using fact-based support systems”

1. Not everybody is data scientist

a. Combining data from multiple sources, b. creating data model and c.authoring reports from it, need special skillset. Many a times, analysts just want an Excel connecting to OLAP cube to do their analysis. Traditional BI has a place in this space.

2. Robust production ready solutions

The analysis done in the analysts R code or ipython notebook is not usually production ready. BI will be very much needed when it needs to be productionised.

3. If ETL is part of BI then its here to stay

How else would you provide quality data to analysts otherwise? And if the definition of BI includes methods to improve process of decision making, then we need BI.

4. Operational reporting is a fact of life

As much as we harp about ad-hoc analysis, data science and self-service BI, plain old operational reporting is a big part of any organization. For that we need classic BI.

SQL Server 2016

SQL Server 2016 was announced in Summer 2015 and CTP is available since then. It comes with some really cool features; JSON support, Polybase and Temporal Tables are some of my favorites. SSRS has seen some much needed face-lift. In all honesty, the UI for SSRS was getting a bit oldish. HTML rendering and Datazen integration are also welcoming improvements. The ability to deploy Power BI desktop reports to SSRS 2016, is also on road map. A very welcome news for organizations who aren’t using Power BI (as in cloud version) yet.

With that in mind, I have created a following presentation (or shall it call sway) which shows all the features I am looking forward to and msdn link for more information. Hope you find it useful too.

Please note that this is not an extensive list of features; they are the features I am interested in.

https://sway.com/s/VwzKzCxF1UJTmNhX/embed

Knight’s Move Puzzle in TSQL

A colleague at work set us an interesting puzzle. The puzzle is a generic programming puzzle but we were required to solve it with TSQL. The person, who can come up with fastest solution, wins. Here’s the puzzle in question.

Puzzle


Imagine a keypad like below.

Alt

We want to find all of the unique 10 key sequences that can be entered into the keypad.

  • The initial key press can be any of the keys.
  • Each subsequent key press must be a knight move from the previous key press.
  • A key can be used any number of times.
  • There can be at most two vowels in each sequence.

A valid knight move is either

  • One step vertically and two steps horizontally
  • One step horizontally and two steps vertically

Examples below of valid moves, (from black to green)

Alt

The SQL Code must

  • Display the total number of possible sequences
  • Store all of the sequences in a Temp table named #Results
  • Not require any permission other than Public and connect on the database that is connected to the window we run the script in

The person who’s code runs fastest wins.

Full disclaimer: I did not win the challenge. A couple of co-workers developed really innovative solutions but I thought it was worth to put my approach in the blog so that I can comeback to it later.
I took- probably – a very simplistic approach and was fairly happy with the results. But that’s not the end of it. When I ran the same solution script on various SQL Server editions, the timings varied widely. They were getting better with newer versions of SQL Server.

Solution


The complete code from below is available on my  github account.

Convert to table

Let’s get the obvious out of the way – this was meant to be solved using TSQL so procedural approach was out of question. TSQL is, as we all know, excells at relational operations and we will get fastest solution if we exploit that. Next up, we want to use the other good part about RDBMS’ in general – the ability to index and do fast lookups.

With that in mind,lets model the keypad as a table. Each alphabet is unique – so that’s our primary key – and each has a set of properties: row location, column location and if it’s vowel or not. This vowel column comes handy later as you will see. The keypad table looks like below.

Alt

Knight’s moves as SQL query

When a knight moves, it arrives at a new location on keypad. This new location has a row number and column number and they can be derived from row and column number of original location and the directions of the move. There are essentially eight cases (or moves) from each character.

One step vertically and two steps horizontally
(1) One step up and two steps on left
(2) One step up and two steps on right
(3) One step down and two steps on left
(4) One step down and two steps on right

One step horizontally and two steps vertically
(5) One step left and two steps up
(6) One step left and two steps down
(7) One step right and two steps up
(8) One step right and two steps up

For each case above, we can join the keypad table to itself on row and column number to find all valid combinations. Lets take case 1 above and apply it to position of H. H is at row 2 and column 3. Lets denote it as H(2,3). One step above H is C and then two steps to the left is A which is A(1,1). So if we reduce row number by 1 and column number by 2 from H, we would arrive at A using a first valid knight’s move. That, essentially is our self join condition. In the same fashion,for case 2 above, to go from H(2,3)to E(1,5) we would reduce a row number but increase the column number by 2 since we are moving to the right. Here’s the sample query for first case

SELECT d1.val,d2.val AS c1_v1up_h2left,CASE
WHEN d2.val IN ( 'A', 'E', 'I', 'O' )
THEN 1
ELSE 0
END isnextvowel
FROM #keypad d1
INNER JOIN #keypad d2
ON d1.rownum - 1 = d2.rownum
AND d2.colnum = d1.colnum - 2

We can apply the same reasoning to all eight cases and that give us eight different queries. The union of all eight queries will be our set of all combinations of valid moves from any character from keypad. Also note that the inner join will automatically eliminate positions which does not exists. For e.g. for 1(4,2) there is no case 3 from above since there isn’t any row below one. Have we used left join, we would have received a record for this case but next val would have been empty. Below are the valid moves from 1.

Alt

Ten sequence chain

Next, we want to find a sequence of ten moves which contain less than two vowels. This is where the vowel column from the keypad table comes handy. But first lets look at how we find the sequence.
We have table at hand which contains every possible valid move from every character in the keypad. Let’s assume that we want to find only three character sequence. We start with our keypad table and then join to valid combination table on current value. So next value column from the valid combination table will be our second key from sequence. If we now join the valid combinations table to results from above on we get the third value in sequence. The join condition would be on next valid character from first result to original value column from second. We can repeat the process to get 10 key sequence.

Here’s the sample query which finds the first three characters

SELECT d.val s1,r1.nextval s2,r2.nextval s3
FROM #keypad d
INNER JOIN #validcombination r1
ON d.val = r1.val
INNER JOIN #validcombination r2
ON r1.nextval = r2.val

Removing vowels

In the ten joins query like above, each instance of the table has vowel column available. If we sum the values of this column in the result set, that tells us how many vowels are there in the sequence. So to filter them out, we simply put in a where clause which checks that this sum is less than two. Using mathematical operations is better than string comparison so SUM is more efficient than checking if more than two characters are vowels. Here are few records from results; you can verify yourself.

Alt

Timings

This is where it gets interesting. I ran the same script on four different versions of SQL Server: SQL Server 2008 R2 SP3, SQL Server 2012 SP1, SQL Server 2014 and SQL Server 2016 CTP 3.2. Admittedly, they ran on different hardware and SQL Server 2014 and SQL Server 2016 CTP was running in a virtual machine but hardware was similar; each had a four processor cores, 8 GB memory and SSD. In case of VMs the configuration exactly same and VM itself was hosted on SSD. I did expect the timings to change but the change was pretty stark. The results are given below. The first run was on a cold cache and subsequent ones are on warm cache. The average column at the end is the average of all.

SQL Version Run 1 Run 2 Run 3 Run 4 Run 5 Average
SQL Server 2008R2 1316 1310 1313 1333 1303 1315
SQL Server 2012 1026 1066 1106 1120 1056 1074.8
SQL Server 2014 403 980 396 373 393 509
SQL Server 2016 470 453 400 390 370 416.6

My theory on improvements in execution timings is that a. The way temporary tables are used in TSQL has been improved in each version of SQL Server and b. The SQL Server query optimizer has been improving with each version. The second point is more relevant since we know that optimizer is improved which each version of SQL Server. I did not check the query plans for each version but may be the answer lies there.

Further improvements to the solution

The code can certainly be improved, although, I believe that unless the approach is radically altered, the gains in improvements will be marginal. It will follow the law of diminishing marginal performance improvements (I completely made that up) whereby the amount of effort to improve performance will grow exponentially as compared to marginal performance benefit we get or, in other words, we will have to put a significant amount of efforts to improve the performance by a littl e. Having said that

  • Putting normal clustered and non clustered index can have an impact
  • How would actual table, instead of temporary tables, perform?
  • In SQL Server 2012 and onward, we can use columnstore index, would they have any impact? Which one would be better, non clustered columnstore or clustered?
  • What if we create the temporary table as in-memory tables in SQL Server 2014 and 2016, would they improve performance?
  • Any combination of the above?

Creating a new SSIS package? Have you thought about these things?

Creating a package in SSIS is easy but creating a “good” SSIS package is a different story. As developers, we tend to jump right into building and creating that wonderfully simple package and often overlook the nitty-gritties. Being an avid developer myself, I must confess that I have fallen prey to this from time to time. When I find myself in a rush to create “that” package, I take a step back and ask myself “have you thought about these things?”. The list below is not comprehensive but it’s a good starting point. I will hopefully keep it updated as I come across more issues. Please note that it is not a SSIS best practices or SSIS optimization tips; these are more of high level things which I keep at the back of my mind whenever I am creating a new SSIS package.

1. SSIS Configuration
Configuration has to be the No.1 thing to think about when you start creating any SSIS package. It governs how the package will run in multiple environments so it is absolutely necessary to pay particular attention to configuration. Some of the questions to ask yourself are

  • Where is configuration stored? Is it XML, dtcConfig file, SQL Server?
  • How easy it is to change configured values?
  • What values are you going to store in configuration? Is it just connection manager or should you be storing any variable values as well.
  • What will happen if the package does not find a particular configured item? Would it fail? Would it do something it should not be doing?
  • How are you storing the passwords if there are any?
  • and so on..

2. SSIS Logging

There is certain minimum information each SSIS package must log. It is not only a good practice but it will make life much easier when package fails in production and you want to know where and why it failed. As rule of thumb, I think the following should be logged

  • The start and end of the package
  • The start and end of each task
  • Any errors on tasks. You should log as much information as possible about the error such as the error message, variable values when the error occurred, server names, file names under processing etc.
  • Row counts in the data flow

Which log provider to use is entirely up to you although I tend to create the log in a SQL Server database because it is easier to query that way.

3. Package restartability

Can you re-run the package as many times as you want? What if the package fails in between the operation? Would it start from where it failed? If it starts from the beginning what would it do?

4. Is your package atomic?

By “atomic”, I mean is it doing just one operation like “load date” or “load customer” or is it doing multiple operations like “load date and update fact”. It is always a good idea to keep packages atomic. This helps in restartability besides helping while debugging the ETL. If you think your package is doing multiple operations in one go, split it into multiple packages.

5. Are you using the correct SSIS tasks?

There are tasks in which SSIS is good at and there are tasks in which databases are good at. For example, databases are good at JOIN operations whereas SSIS can connect to an FTP site with ease. Are you using the optimum task? Can your current operation be done in pure TSQL? If yes, push it to the database.

6. Are you using event handlers?

Event handlers are great if you want to take alternative actions on certain events. For example, if the package fails, OnError event handler can be used to reset tables or notify somebody.

7. Have you thought about data source?

How are you getting data from data source? Is it the best way? Can you add a layer of abstraction between data source and your SSIS package? If you are reading from a relational database, can you create views on it rather than hard-coded SQL queries? If you are reading from flat files, have you set the data type correctly?

8. Naming convention

Is your package aptly named? Does it do what it says on the tin? Does it convey meaningful information about what the package is doing?

Same rules also apply to variables in the package.

9. SSIS Task Names

Have you renamed SSIS Tasks and are they descriptive enough to convey meaning of the operation they are performing?

10. Documentation/Annotations

Is your package well documented? Does it describe WHY it is doing something rather than WHAT it is doing? The former is considered a good documentation although in case of SSIS I find that even the later is very helpful because any new person doesn’t have to go through the package to understand what it is doing. SSIS annotations are great for in-package documentation and can be used effectively.

11. Is your package well structured both operationally and visually?

Can you box tasks into a series sequence containers? Does your package looks like a nice flow either from top-to-bottom or left-to-right? Are there any tasks which are hidden beneath other? Can the person looking at the package for first time grasp what’s happening without digging into each task?

12. Are you using an ETL framwork?

Having a generalized ETL framework will save a significant amount of time because many of the repetitive tasks such as logging, event handlers, configuration can be created in one “template” package and all other packages can be based on this template package.

Please leave a comment about what you think and if there is anything that you always keep in my mind when developing in SSIS. An older post of mine about things to be aware of while developing SSAS cubes is one of favorite and I can see this becoming one too!!

Data Virtualization in Business Intelligence

A long time ago I wrote a blog post where I described three approaches to providing operational reports and compared them against each other. They are 1) traditional DW approach, 2) the abstract views approach and 3) the direct query approach. The blog post can be found here and I keep it handy whenever I am asked about operational reports.

I have been recently looking at Data Virtualization (DV) and started thinking how it can be used in a BI project. This blog is about that. If you are not familiar with DV, this youtube video from one of the DV software vendors provides good introduction. I would highly recommend watching it before proceeding.

This is the wikipedia definition of DV but in a nutshell it is a technique in which multiple data sources are joined together in a logical layer so as to abstract the complexities of the data sources and provide a unified view of the data to the end users. Going back to my original blog post, one of the difficulties of the views approach is in integrating multiple data sources. That’s primarily because views are simply projections on tables in databases. It’s quite difficult to create a view which spans across different databases for e.g. SQL Server view based on data in Oracle unless you some how bring it in the SQL server AKA ETL it. It is even more challenging if the data source is a flat file or XML or JSON. How would you create views on them?

Enter DV. In DV, as mentioned earlier, multiple data sources are joined together in a virtual layer. The type of the data sources can vary from relational databases to files in Hadoop to web services. The ETL is performed on the fly and in-place,if need be, i.e. multiple data sources are integrated in real time. If we can point our reporting tool to this virtual layer, we can provide real time operational reporting. Sweet!! So imagine a quick report you want to knock together which involves SQL Server, a web service and Excel file. DV will gladly connect to three, allow you mash them together and publish this data. You can point the reporting tool to published data and are good to go. And all in real time – that’s added bonus.

This all looks very cool hence a word of caution. It sounds like DV is replacement for ETL and data warehouse but it is NOT. There are far more advantages to having a DW and ETL process in place but they are beyond the scope of this blog. We can certainly think of creative ways of using DV in conjunction with traditional DW and ETL. Providing real time operational reports in a way mentioned above can be one of them.

Estimating SQL Server Table Sizes

LichtenBytes

Microsoft gives guidance for calculating the estimated sizes of your tables and indexes at http://msdn.microsoft.com/en-us/library/ms175991.aspx. Performing such estimates is an important step in determining required server capacity.

Unfortunately, Microsoft’s guidance consists of long textual descriptions of the steps that need to be followed to perform the estimations. It takes some time to interpret the instructions given by Microsoft and perform the calculations.

With that in mind, I put together a spreadsheet to help me perform the calculations. To use the spreadsheet, I simply fill out the spreadsheet with the total number of rows in the table and details of the columns (data types and sizes) in the table and/or index. In my experience, the estimates provided by the spreadsheet are within 5-10% of the actual sizes of the tables/indexes, which has been good enough for my needs.

You can download the spreadsheet here.

To complete the spreadsheet, simply…

View original post 221 more words