SAP Analytics
Thought Leadership

Have you ever wondered if there’s a straightforward way to make your SAP BPC system perform better? The Darwin Calculation Engine (DarCE) might be just the trick you need!

Google “EPM Performance or Memory issues” and you will receive millions of hits from your search results.  When it comes to performance and especially memory consumption, not only is the “magic bullet” fix rare but the resolution is potentially complex and multifaceted in nature.  On a recent project, we started to see symptoms that became issues including long refresh times.  For one user, we saw an Input Schedule consume so much memory it resulted in the “White Screen of Death (WSoD) and Excel crashing.  In this article, I wanted to look at the challenges we faced and strategies employed to resolve the issues.

I was recently working at an organisation where the BPC implementation had developed over time with input
 from a number of third party sources. There were no problems with functionality but, like most clients, they would appreciate any changes which could produce an improvement in performance. Clearly, a variety of skillsets, approaches, and quality standards had produced a challenging situation for this client, and they asked our team to make suggestions.

If you follow best practices for SAP implementation and use, whether it be BPC, Disclosure Management, ECC,
HANA, ABAP, or many others, you will be successful in quantifying the benefits from the suite of products you choose.  Our perspective is that there are no inherently “bad” SAP products, just poor implementation that can make your product use more stressful than it needs to be.

Need to calculate a grand total dynamically?  Shifting of Local Members giving you a headache?  I previously used Indirect() functions, but I discovered with larger reports this put a drag on reporting performance.  To sum from the first report column to the last report column I was using something like this  =SUM(EPMPOSITION(1):INDIRECT(ADDRESS(ROW(),COLUMN()-1))), which is not considered best practice but I got results!

The SAP definition: “EPMALLMEMBERS is used to perform a calculation (such as sum or average) on all the members displayed in a row or column axis, whether the axis contains one or several dimensions”

HOW?  Here are some examples:

Column5 and I have seen a lot of things over the years in the EPM space. I have been working on the technical side of BPC MS since 2000, BPC NW since 2007, and BPC NW on HANA since 2012, having implemented it around the world over 1000 times.

I have a dream to, at some point, write a book of everything I have seen over the years. If investors knew what was really going on deep within some of the world’s datacenters, I think some heads would be rolling! I won’t mention ANY NAMEs but some of them are the who’s who of worldwide companies!

I keep running into the same thing over and over and over again, whether it is hosted by a 3rd party or implemented on-site at a client. The challenges start with a common element that is surprisingly easy to fix: No one seems to take care of the servers and technology!

The Battle of BPC

I have been in the Analytics space for over 13 years, having started at the original OutlookSoft Corporation (creator of the BPC software) as employee #33 (or so). I stayed with the company that acquired OutlookSoft in 2007, SAP, until 2008, when I came to Column5. I have been involved in over 1000 technical implementations of the BPC software (all technology stacks) over the years with companies around the world.

When I say “Technical Implementation”, I mean performing a System Landscape Design (SLD), multiple Installations, Performance Tuning, Disaster Recovery, Environment Synchronization (Transports), and Go-Live Support. As I started the Technical team at Column5 back in 2008 after working at SAP/OutlookSoft for 8 years Column5, has used the above services and others to ensure client success, or as I like to say “EPM NIRVANA”.

Obviously SAP has moved the software forward leaps and bounds in a lot of functional ways, but especially supporting multiple server technology stacks on the backend. This allows SAP to really fit into just about all of the worldwide business technology strategies. Currently with BPC, there are three different “flavors” of the software and they are:

• BPC (10) BW powered by HANA
• BPC (10) NetWeaver
• BPC (10) Microsoft

Sprinkled throughout many books, project kick-off decks, and lists of project success criteria, is the notion that your program or project needs Visible Executive Sponsorship.  If you Google the topic you will see that there is widespread agreement that this is “important”.  You will also see that there are various studies that show that project success is tied to visible Executive Sponsorship and that very high percentages of failed projects list the lack of Executive Sponsorship as a key reason.  However, I rarely see the more practical aspects of recruiting Executive Sponsors, what visibility really means, and tangible advice on how to make the sponsors visible.  Whether you are planning your EPM Strategic Roadmap, or just upgrading from BPC v7.5 to 10.0, getting the right Executive Sponsor on board is essential.

Analytics?  Analytics.  Analytics!  Let us simply get it out there – analytics are the newest, hottest technology that everyone is talking about.  “The CFO’s Technology Imperatives for 2015” by Financial Executives International (FEI) and Gartner states, “many IT organizations have made initial investments, but these tend to be tactically focused and do not address the more fundamental issues of data quality and consistency.”  The market is in such a fervor over the promise of analytics that we often lose sight of our greater IT goals, and we do not effectively outline a strategy that can move us toward the new promised land.  

Surveys from Gartner reaffirm these assertions in a study on Top Big Data Challenges.  The top three challenges include how to derive value, how to define strategy, and how to obtain the skills and capabilities needed related to a big data initiative.

If you have not read part one of this blog, read it here!