Understaning archiving statistics

[former_member555453]

I just did an archiving of financial documents (FI_DOCUMNT) and noticed something with the archiving statistics:

Written DB Space in MB:            473.889

Written DB Table Space in MB:   290.035

Written DB Index Space in MB:     54.059

Written Cluster Space in MB:      129.795

Deleted DB Space in MB:           355.990

Deleted DB Table Space in MB:  297.676

Deleted DB Index Space in MB:  58.314

In other archive tests I have done, the Written and Deleted Space were the same.

How do we explain the difference between the two?

I was just curious why they would be different and how to explain this to someone non-technical.

Thanks.

Mike

[Former Member]

Hi,

As an answer to your question I can provide you some details how

data archiving statistics are calculated starting from release 4.70

(Please find your answer from below description):

How Is the Database Space Information Calculated?

In write programs, the Archive Development Kit (ADK) automatically

collects statistics. In delete, read, and reload programs, you must use

ARCHIVE_GIVE_STATISTICS to pass statistical information to ADK. The

statistics collected by ARCHIVE_GIVE_STATISTICS can be displayed as

described in Statistics.

ADK calculates the database space according to the relevant ABAP Dictionary object type and uses the function module DDIF_TABL_GET to get the width of the relevant object.
You should note, however, that this width is not necessarily identical with the database space occupied by the object. If, for example, you are using a Unicode database with UTF-16 coding, then, n characters (width =n) in a character-type field occupy at least 2n bytes in the database.

After calculating the database space, ADK assigns each result to one of the following database space categories. These categories correspond to the categories in the Statistics function in Archive Administration. The DB Space is the sum of these four categories.

Please note that this information is an extract from the online documenation (section How Is the Database Space Information Calculated?
in the documentation for the 'Archive Development Kit' ->'Developing Archiving Solutions').

Regarding special treatment of cluster tables you can note 565343 (Statistics: Cluster, pool ....).

The DB space values calcuated for cluster tables is diveded by 5 in order to take into account the compression of cluster tables on the data base. It is better to estimate using the number of lines in the logical tables contained in the cluster (for example BSEG instead of RFBLG).

Hopefully this answer is of some help.

Regards

Ruby.

[alper_somuncu]

Hi Michael,

Written DB space in MB = Written (DB Table Space in MB + DB Index Space in MB + Cluster Space in DB)

With same logic, Deleted DB Space in MB = Deleted DB Table Space in MB + Deleted DB Index Space in MB.

Most of the time the archived data (including all DB links and indexes) can be removed after the archiving has been completed successfully but in some cases it is not possible to remove all related DB indexes and links due to the dependencies with other objects / tables.

Because the main target of data archiving is to increase DB performance (not cleaning up some DB space), it's better to analyse and focus on the performance improvement rather than the space cleaned up.

Kind regards,

Alper