SAP business suite on HANA: Businesses must answer the “so what?” question.
So what is Suite on HANA?
SAP announced that their business suite (except SRM) will now be available on their in-memory platform, HANA (SRM will follow soon). This is a follow-up to their launch of the HANA platform about 18 months ago which has primarily been used to support databases in-memory and their analytics platform, Business Warehouse. I was fortunate to be invited to the global press event in Palo Alto, courtesy of Mark Finnern and the SAP Mentor program (Disclosure: SAP covered my travel costs to the event).
Several people have written excellent blogs about Suite on HANA so I will point you to some of these blogs for an explanation of the technology and their implications.
(note: some of the more technical aspects are beyond me! So the technical explanations below are very general and prone to errors (I am sure someone will correct me…)
Key technical aspects
- Everything is in-memory: the biggest bottleneck to speed is the movement of data from an external hard drive (slow, mechanical device) to memory (fast, electronic device). Getting rid of the external drive (except for backups) and putting everything in memory is a radical idea made possible by advancements in technology. This makes SAP HANA fast!
- Leveraging massively parallel processing: hardware (multi-core processors and memory) can be scaled by adding “blades to a rack” as needed. The software has been optimized to take advantage of parallel processing across multiple, multi-core processers. This makes SAP HANA fast!
- Leveraging the capabilities of SQL: typically data are moved from the database to the application where the data are manipulated as needed. SAP HANA pushes a lot of the complex arithmetic into the database so that only the results are moved to the application. Not a radically new concept (stored procedures?), but it is one of the key characteristics of HANA. One related outcome is that (some of the) code is now optimized for HANA (but will still run on non-HANA environments). This makes SAP HANA fast! See Thorsten Franz’s blog: SAP on HANA, and Pushdown for All: News about ABAP’s Adventurous Relationship with the Database for a more detailed discussion of code optimization and code pushdown)
So you get the idea: Innovations in in-memory technology makes SAP HANA is fast!
SAP HANA is fast. So what?
Fast is good. But just fast is not enough! This has always been true with technology and systems innovation. It not enough to simply apply innovative technology to the existing process to make it faster because (a) you are going to make the weaknesses of the process more pronounced, and (b) you are not taking full advantage of what the technology has to offer – what new things you can do with speed. To fully benefit from the innovation, it is essential to re-think business process. We heard several statements along these lines at the press event. SAP HANA can:
- Solve problems that were previously deemed unsolvable
- Develop new business capabilities that were previously not possible
- Require a rethinking of user interactions
Getting back to the “so what” question, here are some implications of the in-memory capabilities of SAP HANA:
- Cost savings: Faster means less time spent on tasks, which can lead to less cost. Other aspects of SAP HANA that were touted to reduce cost are landscape simplification and less code (as some of the work is pushed down to the database).
- OLAP and OLTP in one: traditionally, various enterprise systems (eg., ERP, SCM, BW, etc.) have different databases that are optimized for the type of processing each did (transactional vs. analytic) resulting in a need to constantly move data from transactional systems to analytic system for reporting. Regardless of how often the data are moved, the analytics is always on “old” data. Putting all the data in one place makes for “real” real-time access to data for decision making (hasn’t the meaning of real-time has changed over the years with technology innovation?). Note that today, each system still needs its own HANA environment, but it is not a stretch to imagine that in the very near future there will be only one database. Given that the separate databases are in HANA (in-memory), moving the data can be more real-time than was possible before. What does this mean?
- Instant Analytics and Business Intelligence on transaction data!
- Live conversations with the data rather than depending on pre-defined reports and analytics based on pre-aggregated data. According toHasso Plattner we should be able to converse with data much like we converse with Google: ask a question, get a response, learn from the response, ask another question, and repeat, until you find what you are looking for.
- The end of batch processing! For large organizations there are many tasks that take hours or days to complete. For example, MRP, Dunning, Scheduling, Financial Closing, etc. These can now be interactive!
But these observations only address the “so what” question to a certain point. Leo Araujo wrote a blog on the business value of the Suite on HANA that made the point that the business value proposition must be identified for organizations to realize the benefits of HANA.
- So what if we have instant BI capabilities on transaction data? How can this improve decision making? Managers are used to not-so-current data and have developed processes and procedures to “make do” with what they have. They have to figure out what they can now do that was not possible before – easier said than done. Not just what they can do faster. Rather what kinds of problems can they now begin to address – do they even know what these problems are? How should solutions be implemented?
- So what if a business can run MRP in real time? They are accustomed to running it overnight or twice a day, or whatever. What will running MRP in real time get them that they did not have before? How will this then influence existing business processes? Do they even know what the possibilities are?
- So what if you can have a live conversation with the data? Does a user know what questions to ask and how learn from the mistakes/answers?
You get the point. There will be change. There must be change – in processes and user interactions – to truly benefit from the Suite on HANA.
Transformation without disruption.
One interesting point that was repeated many times during the press event is “transformation without disruption.” Is that even possible? I heard several people ask that question in conversations at the press event. Upon reflection, it seems to me that the SAP executives were really talking about the technical transformation rather than business transformation. That is, the transformation from the non-HANA to the HANA technical platform will be non-disruptive. The technical migration involves migration of the database to the HANA database and applying a service pack to the system. I will let the technical experts comment on how non-disruptive this will be! But can business transformation be non-disruptive? I am not so sure. Changing how you interrogate data to make decisions and solve problems that were previously unsolvable, and taking advantage of doing something in real time that took hours or days will require a disruption in how users think of their processes and how they interact with the system.
When asked about how businesses will understand the business value of the Suite on HANA, SAP executives responded with: use design thinking. In fact, one statement was that SAP would send design thinking experts to companies and, in about 24 hours, help them understand the business value. Perhaps it will be that simple. I think not.
Design thinking essentially says the user must be the focus of all design. Most of the activities in the design thinking process are aimed at understanding the specific needs of a user and creating a design to meet these needs. Is this something new? I think not! Those of us in the Management Information Systems discipline (primarily in B-schools) have advocated and taught user-centric design and the importannce of user involvement in design for decades. I am glad that engineers and computer scientists have now become aware of the importance of focusing on the end user 😉 . Perhaps, finally, finally, SAPGUI will be history. I am an optimist.
Training and Education
In a comment his blog noted above, Leo asks who is responsible for educating businesses and users. There are several stakeholders. The businesses themselves must want to learn. SAP must provide clear examples and use cases that demonstrate the business value of SAP HANA. Consultants will play a major role. And, for the next generation of leaders, the SAP University Alliances program will have a huge role. I am sure, more will be written on exactly what the role of these stakeholders is.
One thing that troubled me is the reaction of some of the executives at the press event to the question of what is SAP doing to provide adequate training and education to users and businesses. The responses by some executives were something to the effect: “if we need to provide training then we have failed” or “how much training did you need to use your smartphone?” or “we have made it so easy that one will not need training.” Yes, SAP has created some cool user interfaces recently: the 10 or so dashboards for the Suite; SAP Personas; interfaces for mobile apps, etc. These are simple and easy to use and perhaps (although I am not convinced) they will not need user training. However, most of SAP users are still on SAPGUI! This has not changed with Suite on HANA. Further, and more importantly, it is not so much about training a user on how to use a particular interface; it is about educating users and organizations about what they can do with the technology – the value they can gain for themselves and their organizations. Nothing the SAP does will eliminate the need for this type of education. Design thinking is not just about user-centric design; it is also about ensuring that the user understands the value of the system and actually uses it.