HANA is considered to be an ideal platform for applications in the cloud. It consists of certain unique features and is a good choice for software service. This also applies to SERP on HEC, which are the business network applications, the customer as well as partner applications on HCP as well as many new services from the startups.
Applications are offered as a service promise on an instant availability, easy configuration, elasticity that grows from 5 to 500 users, continuous improvement, and connectivity too many other cloud services, guaranteed service level, a hassle-free operation at the lower costs. The service provider has to cope up with these challenges by building the applications in the right way that really focus on ease of its use, choosing the hardware and the system software carefully, keeping the above-mentioned goals in mind.
The services in the respective SAP Cloud also vary from marketplace (business network) applications such as ARIBA, Fieldglass, and Concur to enterprise-specific applications like SERP, B1, BYD or even Success Factors. The given underlying database has to in fact handle the large data sets and many smaller ones at a low cost and rather good performance.
HANA is accepted as an in-memory database with the mostly columnar store. Along with the new developments in hardware, where larger amounts of memory can be shared by many of the CPUs with up to 18 cores each (INTEL Haswell) as it offers some very interesting features, that are especially helpful in the respective cloud such as :
- The columnar store architecture permits the use of massive parallelism.
- The dictionary-based compression (>5x) does reduce the data footprint for all sorts of applications.
- The data is being actually converted into the integer, which speeds up the required database internal processing.
- Columns of tables, which are not usually being populated, are not just taking any space. This feature is also important for the generic applications thus serving a variety of industries as well as locations.
- New attributes can no doubt be added on the fly without disrupting the cloud services.All columns of a table are able to work as an index.
- The speed of HANA does make older concepts such as transactional aggregation, data cube roll up, materialized projections or views superfluous and thus accelerates the data entry transactions
- For the respective historical data, HANA does scale out of the approach.
- HANA incorporates the text and geospatial capabilities.
- Sophisticated libraries for business functions or even planning-forecasting-and-simulation algorithms help simplify the application code.
The combination of OLTP and OLAP in one single database.
- HANA is fully compatible with Oracle, IBM DB2, MS SQL Server, and SAP ASE, only does store procedures that have to be translated or rewritten for legal reasons.
- Multi-tenancy is via the database for smaller application deployments or generic applications of the business network type that will l come soon.
- With the help of HANA the percentage of updates of all database activities are minimal and thus the remaining ones are done in insert-only mode, which nearly does nearly eliminate the need for database locks rather completely.
- A very extreme high-speed version of HANA, with a very small footprint for the deployment of data marts or even on frontend clients, will be shipped very soon.
- The architecture of the data store does support the caching of intermediate results, while the new data input does add on automatically.
A separate data warehouse, running only of course OLAP type applications is still valuable and well supported by HANA.
All these features do make HANA a very attractive and efficient database for cloud-based applications of any sort.