Adaptive Computing  

Make Your System Landscape More Flexible

A system landscape usually consists of a number of SAP systems of different types, different usage, and with different loads – in a hardware park that itself consists of servers of different sizes with different capacities. Conservative designs of a system and landscape architecture are still very static. Once a solution has been installed at a customer site and production operation has started, an IT landscape of this type is usually only significantly modified during structural changes such as upgrade, enhancements, or migrations. Constant adjustment to the continuously changing requirements – for example, with regard to performance – can only be implemented to a limited extent with these resources. This situation is therefore suitable for a modernization that brings flexibility, uses resources optimally, contributes to landscape consolidation, makes operation easier, and consequently reduces costs.

 

The objective of Adaptive Computing is to operate “any server – any time – anywhere”.

 

Purpose

Adaptive Computing (AC) offers an approach for the dynamic assignment of hardware resources. In an Adaptive Computing Infrastructure, you can, for example, very quickly move a system for which a server is no longer providing the required performance to a more powerful server.  Dialog instances can be flexibly started on free servers to absorb spikes in workload. Another situation in the computer center could be development and training systems that are stopped at night to support the background operation of productive systems with their free servers. These entire processes can be performed without needing to make any configuration changes.

 

Concept

The Adaptive Computing concept starts Application Services at the instance level. An Application Service is the smallest currently installable unit of an SAP system. This can be the central instance, the dialog instance, and the database.

Without Adaptive Computing, it was very difficult in the past to move SAP systems or instances from one host to another. Projects of this type sometimes required extensive reconfiguration of the instances. With Adaptive Computing, it is now possible to move an Application Service from one host to another in a few minutes. This time consists more-or-less only of the time required to stop and restart the Application Service.

The aim of Adaptive Computing is to replace the previously inflexible link between hardware and SAP system with the concept of a central storage medium and resource sharing. Resource sharing is enabled by organizing the system landscape into Building Blocks, which allows the system landscape to react extremely flexibly to current demands. All servers, whether physical or virtual, are designated as resources, which can in turn be logically grouped and separated into pools. All resources within a pool are linked with each other by their own network.