A delta merge is used to transfer modifications from the delta storage to the main storage.
First, an asynchronous check is performed to see whether a delta merge is required. If a
threshold value is exceeded, the merge is carried out in the delta storage. When a read access
is executed, the data is read from the main storage and the delta storage and the results are
merged together. Due to the fact that SAP BW normally processes mass data, the delta
merge is also managed by the application in this case. The ways of working depends on the
object type (InfoProvider). The delta merge is either performed automatically by the
application, or must be triggered manually within the application:
Column-store tables are good at reading performance, but poor at writing or receiving new,
updated data. For this reason, in-memory column-store tables consist of a main storage
(read-optimized), as well as a small delta storage (write-optimized). When reading such a
table, both the main and delta parts are relevant. This split is fully transparent for the
application and its users, and is managed by the SAP HANA database.
The same is true for the delta merge process. The transfer of data from delta to main is
managed by the database, and is a task performed by the database administrator.
Delta merge is a concept for in-memory column store tables only. It does not
apply to tables managed in the in-memory row-store or tables managed outside of
SAP HANA memory.
There are two ways which can be used for delta merge -
1. Setting in DataTransferProcess : The Data Transfer Process (DTP) has an Update tab that
contains a Trigger Database Merge checkbox. Once the DTP request has been
successfully processed, this setting controls the delta merge. The checkbox is selected by
default.
2. The Trigger Delta Merge process type can be used within a process chain to execute the delta merge.
No comments:
Post a Comment