SAP HANA is a database on which, for example, SAP Business Warehouse or SAP Business Suite can be operated. It is often referred to not as a “database” but as the “SAP HANA platform”.
In the SAP S/4HANA solution, SAP’s own database SAP HANA is an essential part of the solution.
What are the advantages of SAP HANA?
- SAP can tailor the software to database technology
- Application and database layers are more closely connected
- Only one database technology understood and worked out
If you want to run an SAP HANA platform, you need servers based on either Intel x86 or IBM Power architecture. There are basically two ways to implement this: SAP HANA appliances and Tailored Data-center Integration (TDI).
SAP HANA Appliances are complete systems provided by SAP’s hardware partners as certified solutions. These solutions already contain all the components needed to successfully install and operate the platform. They are easy to install and can be made ready for use quickly. The appliances are available in different sizes to meet different requirements.
However, if you have specific platform requirements or want to leverage existing hardware, TDI may be the better option for you. With TDI, you have the ability to customize the individual components such as storage, virtualization and networking. You also have more flexibility in sizing and can customize the platform to meet your specific needs.
Our team of experienced SAP consultants is available to help you choose the right option and to help you install and operate your SAP HANA platform.
Justin Tönjes, CEO of Tönjes Consulting GmbH
Want to learn more about SAP HANA and its implementation?
SAP HANA can also be operated as a single-host system (scale-up scenario) or as a multi-host system (scale-out scenario). While the SAP HANA platform is operated on one host in the scale-up scenario, it is distributed across multiple hosts in the scale-out scenario. This is useful, for example, to circumvent the physical limits of individual hosts in terms of main memory and/or the number of CPUs. Currently, a scale-up system can manage a maximum of 24 tebibytes (TiB) of RAM, while scale-out systems can realize significantly larger dimensions. However, in the scale-out scenario, complexity and internal management effort also increase.
Most SAP S/4HANA systems are designed as scale-up systems, which means they run on a single large machine. Only in cases of very large systems with a required RAM of at least 6 TB is operation as a scale-out system permitted.
Operating systems for SAP S/4HANA systems are limited to Suse Linux Enterprise Server (SLES) or Red Hat Enterprise Linux (RHEL). You must ensure that you use a compatible version of the operating system for your SAP HANA version.
Architecture
SAP HANA is a high-performance in-memory database platform that stores a large part of the data to be processed in the main memory of the SAP HANA server. Compared to classic database systems, which can only store a fraction of the data in main memory, SAP HANA offers significantly higher performance.
However, SAP HANA also requires a persistence layer to store the data permanently. This ensures that the data remains available even after the server is restarted. The persistence layer is an important part of the SAP HANA architecture and ensures that data is stored reliably and securely.
Column Store
SAP HANA offers the option of storing data in the column store column-by-column instead of row-by-row in the row store as in conventional relational databases. In modern SAP S/4HANA systems, most data is stored in a column-based manner. This storage type offers several advantages:
| Higher compression rates | By storing data in columns, data of the same type is arranged directly next to each other. SAP HANA can thus select an algorithm for each column that optimally compresses the data. Dictionary compression is also used in the Column Store, whereby different values of a column are numbered consecutively and only need to be stored once in a dictionary. |
| Higher speed for column operations | For queries on individual columns of a table, as is often the case with search or aggregation queries, the column-by-column storage provides a great speed advantage. |
| Waiver of indices | Column-based data storage can be equated with an index on each column of a row-based data storage, making indexes superfluous in most cases. This has positive effects on write speed and memory consumption. |
| Waiver of aggregates | With SAP HANA, aggregates no longer need to be calculated in advance and stored in additional tables, as they can be calculated directly and quickly as needed. This advantage has changed the data model in SAP S/4HANA compared to the classic SAP Business Suite. |
| Parallelization | Column orientation allows database operations on different columns of a table to be distributed across multiple CPU cores. However, data can still be stored row-based in SAP HANA. In certain cases, this can be useful, for example, if there are few rows, if you mainly work with one data set, if there are many different values in the columns, or if neither aggregations nor search operations are required. |
Delta Merge
The decision whether a standard table is stored in Column Store or Row Store has been made by SAP and can be viewed via transaction SE11 in the SAP GUI. For custom tables, it is the responsibility of the development team to choose the correct storage type, with the default being Column Store.
Sizing
Adjusting data volumes in classic databases is comparatively straightforward, as additional storage media can be added. In contrast, SAP HANA’s in-memory technology makes sizing more difficult, as the main memory or even the entire hardware may have to be replaced. Consequently, an exact determination of the required main and CPU memory is crucial for initial sizing, as these components are very expensive and security buffers should not be too large. Depending on whether the SAP S/4HANA system is a new implementation or a conversion of the existing SAP Business Suite, different tools are available for sizing. It is important to factor in the expected data growth for the planned useful life period, as the tools only take the current point in time into account. The sizing report from SAP Note 1872170 can be used not only for planning, but also for checking the actual system size on the SAP S/4HANA system.
Dynamic Tiering
In recent years, SAP has developed various technologies under the term Dynamic Tiering to more cost-effectively store data that does not need to be retrieved quickly in main memory. These include Data Aging and SAP HANA Native Storage Extension (NSE), although Data Aging is now discouraged by SAP. Read more about Dynamic Tiering in our article.
Contact us!