An off-topic spot to chat about your musings of the day

OLAP and Visualization

Reply
Occasional Contributor
Posts: 14

OLAP and Visualization

[ Edited ]

As part of my course curriculum for MS in Analytics, I had an opportunity to discuss OLAP and visualization. I thought of sharing it with SAS community.

 

Online Analytical Processing (OLAP) helps analysts portray data in a multidimensional format, typically in the form of a cube and may include additional graphics i.e., bar charts, line charts or visualization to get the point across to audience, vis-à-vis, consumers of data. OLAP provides slices and dices of data.

 

Example: If the CEO of an organization wants to understand production performance of an organization by:

 

Dimension 1: Time: by month (Jan-Dec), quarter (Q1-Q4), year (2016),

 

Dimension 2: Line of Business (LOB): by laptops, desktops, servers for Dell Corporation, and

 

Dimension 3: Customer Segments: Residential, Commercial, and Defense.

 

Visualization takes a snapshot of past and current states of data in the form of illustrations, pictures, graphs, and charts and provide shape, size, and color to make the data come to life and make these tangible for consumers of data. Through visualization, decision makers could understand overall business impacts and take needed actions that impact business direction.

 

                                           Similarities between OLAP and Visualization

 

                                  OLAP

                         Visualization

Visually displays data in the form of cubes

Visually displays data through illustrations, charts, graphs i.e., histograms, scatter plots, bubble charts and so forth

Communicates data in 3D per above example.

Visualization mostly communicates data in 2D and sometimes through animation, could be illustrations, photos etc. as well.

Analysts have accessibility to manipulate and work with data across different dimensions which updates the cube in a timely fashion

Visualization is a great communication tool to make a point across specifically on trends, patterns and anomalies such as outliers.

 

                                          Differences between OLAP and Visualization

                                  OLAP

                             Visualization

Extracting insights from OLAP might become complex rapidly depending on the number of dimensions and multiple cubes could cause confusion.

Visualization might be easier to understand, however the amount of information conveyed might be limited with simple diagrams such as histograms.

The producer and consumer of OLAP cubes might be different which might lead to lack of clarity.

With effective visualization, commonality could be established across different functions of the business.

Updating OLAP requires dedicated resources.

Once set up correctly, the process of visualization could be automated.

 

Reference:

 

Sabherwal, R. & Becerra-Fernandez, I. (2013). Business intelligence. Hoboken, NJ: Wiley. ISBN: 9780470461709.

Community Manager
Posts: 2,842

Re: OLAP and Visualization

Thanks @Murali11!  To listen to the hype, you'd think that OLAP has fallen out of favor thanks to large in-memory visualization techniques and fast computing -- the argument being that there is now less of a need for pre-summarization of dimensions/metrics that support traditional OLAP.

 

However, I know that our SAS customers still use OLAP quite a bit.  The OLAP Analyzer in SAS Enterprise Guide supports Microsoft Analysis Services as well as SAS OLAP server, and lots of people still rely on those.  And the OLAP Analyzer is a real gem -- so powerful, even when working with non-SAS data.

Occasional Contributor
Posts: 14

Re: OLAP and Visualization

Chris,

 

Thank you for your reesponse. I agree. Most of the small to mid-scale organizations might not invest in additional services given the market/economic fluctuations and the year over year organizational performance, as one might imagine. 

Ask a Question
Discussion stats
  • 2 replies
  • 305 views
  • 1 like
  • 2 in conversation