As part of our celebrations for #SAPHANATurns10, I had the chance to talk with Alexander Boehm, Executive Database Expert and Chief Architect, SAP HANA & Analytics. Alexander joined SAP back in 2010 and has played a central role in the evolution of SAP HANA over the past decade. We caught up in September 2020 to talk about his HANA highlights, his favorite features of SAP’s flagship database, and what’s next for him and SAP HANA.
Jessica: What’s your role in the SAP HANA & Analytics team?
Alexander: I’m one of the chief architects for our SAP HANA in-memory database management system. Part of my role is to outline the strategic direction for SAP’s data management offerings and advise our senior management team on decision alternatives, as well as the latest developments in the database community and their possible implications.
In addition, I’m also involved in many operational projects designed to further evolve the architecture and design of SAP HANA database systems in several key areas, such as performance, security, operations, and leveraging the capabilities of modern hardware.
Jessica: When did you join the SAP HANA team?
Alexander: I have been with the SAP HANA team since I joined SAP on April 1, 2010, as an engineer in the newly founded in-memory database team. At this time, my colleagues had just completed the first internal prototype that proved that this new kind of in-memory database was technically feasible. While we didn’t use the term “HANA” back then, the team and product evolved from this initiative into what SAP HANA is today.
Jessica: What made you want to join the SAP HANA team?
Alexander: I was doing a PhD in database management systems at the University of Mannheim, Germany, from 2005 to 2010, and it was clear to me that I wanted to work on database systems afterwards. I was actually preparing to relocate to the US, where most of the database teams were based back then. I was especially interested in the possibilities of modern hardware and how to leverage them for data processing. Then, a former colleague from university mentioned that SAP was recruiting developers for a new, modern, in-memory database management systems, so I submitted an application. Already during the interview, I really liked the discussions and the spirit of the team, and I joined shortly after handing in my dissertation.
Jessica: What is your favorite feature or functionality of SAP HANA?
Alexander: I really like our columnar runtime system, as it is extremely versatile and fast. It is also the foundation of the success of the overall SAP HANA system. On a more detailed level, the components that I personally contributed code to are, of course, still dear to my heart, including our landscape and metadata management components, the repository, as well as some of the smaller features like the table reload or the table consistency check.
Jessica: Do you have any highlights of your time so far working on SAP HANA?
Alexander: In my first year with the team, we were able to successfully use SAP HANA as an analytical appliance and then later as the foundation for SAP Business Warehouse. Motivated by this success, we broadened the scope of the project from analytical use cases to also include transaction processing. Our ultimate goal was to run complete enterprise application stacks on SAP HANA, as envisioned by Prof. Plattner already some years before. Back then, this idea was heavily debated, even in academia, and several renowned experts were skeptical that this kind of hybrid transactional/analytical processing system could become a reality. We had to redesign several components in the system and develop many additional optimizations, but, in the end, the system was able to efficiently run the most demanding enterprise OLTP workloads with the most complex analytical tasks. This proved that the combination of OLTP and OLAP in a single system was no fantasy.
Jessica: Why does this moment stick out to you?
Alexander: From an engineering perspective, we took on a challenge that even the database research community was hesitant to take on and successfully turned it into an enterprise-class database management system. From a business perspective, we were able to define a completely new class of operational database management systems to provide additional value for our customers and applications.
For me, personally, I drove and coordinated the re-architecture and optimization tasks on the SAP HANA side. The process of not only revisiting many of our early design choices and collaborating closely with many of the SAP HANA development experts, but also interacting with many other teams and colleagues at SAP was phenomenal. We had intense discussions with our central performance and scalability teams, many application development groups, field experts, and even internal and external customers. This all helped us immensely when it came to understanding the requirements and reshaping the system.
Jessica: What would you say was the impact of this work in the end?
Alexander: The outcome of this effort was that SAP HANA could run the business suite, SAP’s enterprise resource planning (ERP) flagship. We achieved this in record time as a joint project with many colleagues and teams at SAP, pilot customers, and successful deployments of our own IT systems.
In addition to leveraging the power and performance of SAP HANA for existing SAP solutions, we, as a company, were also able to realize the positive impact a fast and versatile database can have on other scenarios and use cases – first and foremost for the end user of those systems.
On the one hand, this led to the creation of SAP S/4HANA, a completely new ERP solution that is specifically designed to benefit from the strengths of SAP HANA. On the other hand, it was also the starting point of our journey to move our other applications, such as SAP Ariba or SAP SuccessFactors, on SAP HANA instead of third-party databases, which has helped them to substantially improve system response times, capabilities, and the end-user experience.
Jessica: What’s next for you and SAP HANA?
Alexander: Many of our customers are interested in cloud-based data management solutions. This means that, in addition to the continued incremental evolution of the database kernel itself, a major focus area these days is on extending and improving our cloud capabilities. This includes aspects like improved elasticity and availability, as well as making the system more resilient, intelligent and autonomous. In addition to these customer-facing aspects, there are also interesting software development challenges, such as disaggregation and componentization, that are becoming increasingly relevant in large, cloud-scale deployments. All in all, there’s a lot to look forward to over the coming months!
Jessica: Thanks, Alexander, for taking the time to speak to us!
