IT Infrastructure and Analytics

SAP S/4 HANA – OLTP on OLAP system – changing the backoffice

Share this post:

If SAP’s vision of S/4 HANA becomes real with Simple Finance and Simple Logistics being the start of the rewrite of ERP then the S/4 database will become the analytics database that runs all the backoffice transactions of your business with connections to the SaaS based systems and will make real-time business a reality. While the speed of in-memory is attractive, it is the up to date high integrity data with powerful analytics tools that is the true power of S/4 HANA. The ability to understand the DNA of business, the data, and turn it into information is compelling.

Today, we spend 70% of our effort for analytics finding, moving, cleaning, transforming, and validating data to provide analytics answers. We spend vast piles of money to make it work and we are still often out of sync. No better example can be found then most financial close process which may start days or even weeks before to gather the data. So if you gather financial data from systems on 12th through the 24th of that month, you have a close based 12+ days. Worse, if anything changed, such as major sale (positive) or huge purchase (negative), your finances will be inaccurate. The close on the last day of the month is based on data that is 6 to 19 days old from a multitude of sources not synchronized.

Our current ERP ecosystem (simplified).

Our current ERP ecosystem (simplified).

The multi-ledger model is great when you are running a dry goods store with paper. In the computer age, we made it faster, and easier to calculate, but we haven’t changed the process at gross level. We still take lots of “ledgers” which might be for sales such as Point of Sale systems (POS), procurement processes, credit cards, etc. and put them into another ledger we can close. The Universal Ledger, table ACDOCA (see below), in in S/4 HANA changes the process. Everything runs on that single universal ledger giving you a real time view at any point in time. The technology changes the process, not just automates the old process.

The S/4 Universal Ledger (ACDOCA) provides real-time, consolidated view of Finance.

The S/4 Universal Ledger (ACDOCA) provides real-time, consolidated view of Finance.

The Line of Business (LOB) applications like CRM, PLM, SCM, SRM, and few other minor LOB applications, data and processes will fold into S/4 HANA. In addition, S/4 HANA will be designed to connect to SaaS based applications like Ariba, SuccessFactors, etc. to pull that data in as well. As S/4 HANA evolves, it will become the single source, real-time repository for all business operations and back office data moving us towards the single version of the truth we’ve been promised from day 1 of our ERP journey.

The S/4 HANA system integrating LOB applications and SaaS.

The S/4 HANA system integrating LOB applications and SaaS.

Now that all the data is in one location, database, collection, etc. you can use the analtyics capabilities of SAP HANA to analyze the data. Because your data stays where it originates, it has high integrity. The denormalized models provide for simpler view of the data. The HANA analytics tools allow you to powerfully and quickly analyze the data. The S/4 HANA database truly become single source of your internal business data. You are now operating your business on your analytics database.

There are still problems to be solved and questions to answer. I’ve been in IT for a long time, so I know everything is fixed in the next version and the tale of S/4 HANA is laid out in many versions. The functionality is being developed as we speak and it will take some to time to mature. The LOB applications are not yet pulled in. HANA itself is maturing, but does not yet have the proven track record of IBM DB2 or other databases. The S/4 to SaaS options have not evolved as fast or open as they need to become. In general, how fast can SAP pull this together and stabilize it.

About 18 months ago, I would have predicted that on-premise ERP would dissolve into world of SaaS with the remnants going to PaaS custom solutions built using DevOps. Some of ERP will go that way regardless of S/4 HANA’s ability, but S/4 HANA now becomes value add player in running your business worthy of updating to or implementing for the first time. No other solution can enable you to run so much of your transactional business on an analytics database for real answers to your information requirements.

Each company needs to ask itself, when do we need this capability? When will our competition have the capability and get in front of us? Can we gain advantage by investing in S/4 HANA. I think for many companies time to move forward will come rapidly even while the vision is playing out. At IBM, we are seeing an increased number of clients moving forward with S/4 HANA every day and helping make them successful.

Abbreviations:

  • ERP – Enterprise Resource Planning
  • ECC – Enterprise Central Component
  • BW – Business Warehouse (SAP’s data warehouse)
  • CRM – Customer Relationship Management
  • PLM – Product Lifecycle Management
  • SCM – Supply Chain Management (includes APO – Advanced Planning & Optimization)
  • SRM – Supplier Relationship Management
  • SaaS – Software-as-a-Service
  • PaaS – Platform-as-a-Service
  • CEC – Customer Engagement Center (CRM successor at SAP)
  • HCP – HANA Cloud Patform
  • HCI – HANA Cloud Integration
  • SF – SAP SuccessFactors (unofficial abbreviation) (SAP’s SaaS based Human Resources application)
  • GB – Gigabyte (measurement of storage, disk based in these diagrams)

 

Add Comment
6 Comments

Leave a Reply

Your email address will not be published.Required fields are marked *


greg misiorek (@greg_not_so)

Hi Chuck,

nice and succinct overview of what S/4HANA can offer to existing SAP installations. i
just have one clarification question. by “The Universal Ledger”, do you, by any means, mean the ACDOCA table or something else?

thank you,

greg

Reply

    Chuck Kichler

    Greg,
    Yes, you are correct. I am referring to the ACDOCA table as illustrated in this SAP client facing presentation added into the post(above). I could not get the graphic in the reply. I think it will help other readers. Thank you for the compliment.

    Reply

Ying

Hi Chuck,
I am very impressed by your article. We are going to build ECC and CRM on the HANA platform. But I think it should be a operational system instead of an analysis system. The reason why I say it is that OLTP system needs highly normalized data model to store transactional data which is not good for data analyzing. So I am wondering how SAP S/4 HANA can provide 2 different system at the same time. Would you please explain it for me? Thanks in advance for your explanation.
Ying

Reply

    Chuck Kichler

    Hi Ying,
    HANA changes the rules for databases. I started as DBA and normalized all my databases to 3rd normal form. While the data was easy for the computer to related, it was difficult for humans to understand even if you had the full ER diagram. Now with SAP ECC having something like 55K tables, it is a real mess to understand.
    Simplification is done by denormalizing the storage of the information. This makes the data simpler for humans to understand. The in-memory speed then allows you to do the aggregation, analytics, on the fly. You have to go back to why Dr. Codd and others pushed for normalization of databases. It was to avoid duplication and the efficiency of the machine especially with slower storage mechanisms. Now that it is all in memory, we can check for duplication and we can ignore the machine efficiency and focus on the human efficiency.

    Reply

      Ying

      Thanks Chuck for your reply. You are right, normalized models helps integration check and efficiency for machine operations, such as Create, Update and Delete etc.
      But I still have another question for SAP S/4 HANA. You mention that HANA changes the old way to store transaction data with denormalized models, but does it store historical data? In our traditional data warehouse, we extract, transform and load data from source to OLAP system, and may take different SCD(slowly changing dimension) type to store data. Does SAP S/4 HANA has the similar way to store historical data for data analysis?
      Thanks again for your help.

      Ying

      Reply

        Chuck Kichler

        Ying,
        All transactions include time stamps, so the history is stored. Transactional integrity is improved with S/4 HANA. Aggregates are computed in memory, so they don’t need to be prebuilt and stored.

        For BW on HANA, clients are using Near-line Storage (NLS) to age out the data to disk as required as memory becomes pricey. We also are seeing clients starting to integrate HADOOP into HANA as lower cost way to store massive volumes of raw information multiple channels. This is not as applicable for S/4 HANA as it is where BW is being used as “the BI” for the corporation. HANA does integrate with on-premise and cloud based HADOOP solutions. I’m personally following this trend especially now as SPARK, a new in-memory DB technology closely aligned to HADOOP from Apache Software Foundation, has just came into the market. IBM is strongly backing it as part of our overall open standards direction even though these products compete with our own DB2 products.

        Reply
More IT Infrastructure and Analytics Stories

What if you could keep track of customer preferences at the city and neighborhood level or better yet predict them?

When the Brewers are playing in Milwaukee, beer starts flying off the shelves. When the Doo-Dah parade marches in Columbus, Ohio, water pistols suddenly surge. And when the Blue Angels fly over Seattle, sales of sunglasses soar. Each of these local events has a profound impact on the behaviors of the people who live there. […]

Continue reading

Five strategic priorities for digitalization across IM&C businesses

Digitalization has already reached almost every aspect of today’s life and will no doubt embed itself further into our personal and business lives. For the industrial machinery and components (IM&C) industry, new and exciting opportunities are provided by advances in hyper-connectivity, automation, cloud computing, big data and analytics, computing on the edge, machine learning, augmented […]

Continue reading

A case for a business case: SAP S/4HANA

Three years ago, on a chilly Tuesday morning during the winter of 2015, after ringing the opening bell at the New York Stock Exchange (NYSE), SAP announced its next generation S/4HANA ERP solution in a highly orchestrated press conference as the Exchange opened for another day of trading. However, recent figures show that out of […]

Continue reading