What are the PBS archive add ons?
PBS archive add-ons® are add-on solutions that can be used immediately for SAP data archiving. They are tailored to corresponding SAP modules and ensure that you can access all of your data quickly and seamlessly - in both archives and databases simultaneously - even after archiving.

What are the requirements for installing the PBS archive add ons?
You need to have successfully completed your SAP data archiving. Free capacity in the file system of your operating system is also required. The software is installed in the SAP program library using SAP standard tools. The only difference from the SAP standard is a PBS-specific namespace, /PBS/, that must be created first.

In which SAP solutions can PBS software be used?
Most PBS software products are mainly used in SAP ERP or R/3. The PBS archive add ons are tailored to the corresponding SAP modules. PBS provides an expanding range of products for SAP NetWeaver BI, SAP Industry Solutions, and SAP Retail.

For which SAP archiving objects are PBS solutions available?
The development of new PBS solutions is pragmatic and focused on our customers' requirements. We now support most SAP archiving objects. You can access an overview under 'Availability by Archiving Objects'. To optimize your SAP data archiving, we recommend advance implementation of PBS Database Analyzer PLUS to determine the growth rate of your data volume and which modules are particularly affected.

On which platforms can PBS software be used?
PBS software can be used on any platform suitable for running SAP software.

Do I need a specific database license for PBS software?
No. Like SAP archives, the data generated by PBS software is stored in the file system of the operating system. Only about ten bytes per document are stored in the database tables set up during the installation of PBS software in an SAP system.

With which SAP releases can PBS software be used?
Many PBS archive add ons are already available for SAP R/3 3.1I and later. New modules will be developed first for the current SAP release and then ported to earlier SAP releases, if required. You can access an overview under 'Availability of the PBS archive add ons'.

In which languages are PBS solutions available?
German, English, and French are currently supported. For all other languages, we provide a PBS tool with which you can automatically generate translations in the language desired for up to 95 percent of the software.

Is special end-user training necessary?
Since the transactions involved are highly analogous to those in SAP standard software, end users do not need special training. For SAP R/3 4.0 and later, the SAP transaction name is preceded by the namespace /PBS/ (for example, /PBS/FB03 instead of FB03).

Do the PBS archive add ons have a specific authorization concept?
No. The authorization checks are the same as in the corresponding SAP transactions.

The user profile of administrators and users has to be added for indexing and display of archived data. In addition to the general authorization for the transaction namespace "/PBS/", the authorizations S_ARCHIVE and S_DATASET with the activity 01 (Create) have to be added to the profile of administrators. End users need the authorization objects S_ARCHIVE activity 3 for "Display", and S_DATASET with activity 33 for "Read".

How large is the required disk space?
The required disk space decreases significantly due to the SAP data archiving and the subsequent PBS indexing. The data is compressed in both batch runs. Thus compared to the database, savings of up to 90% are possible in the logistics area. The gain in disk space reduces to only about 60% in financial accounting. This is due to the line items already being stored in compressed form in the database, and the savings effect mainly results from the saving of disk space for the index tables BSIS, BSAS, BSAD and BSAK that are reduced again by about 90%.

Measurements at customers have resulted in an archived accounting document including the PBS index data having an average size of about 600 to 700 bytes. An archived SD order including the PBS index data has an average size of about 1500 to 1700 bytes.

What are the consequences for the database due to archiving?
By relieving the load on the R/3 database, the performance and thus the response time of the system increase. Moreover, the times for the system backup decrease. The certainty increases in a large system for being able to execute a restore within an acceptable timeframe. An upgrade to a higher SAP release is often only possible after reducing the data through archiving. You can realize this using the PBS software so that the user does not have to do without a convenient access.

Since the archived data and the PBS indices are outside the database, the disk space in the database, and therefore the costs involved for the administration are reduced considerably. PBS requires only about 10 bytes per document (which is only one GB for 100 million documents) in the database to access the PBS indices. Thus the solutions with the least disk space requirement in the database are the PBS solutions. You also receive the disk space savings in a QA system that is usually a copy of the production system for a key date. You can access the productive SAP archive from both systems using the PBS indexing so that you can test programs that access the archive in the QA system. The disk space for the archives is only required once. If you reduce a database from 400 GB to 200 GB, the QA system also reduces to 200 GB. You have saved 400 GB in the database in which about 50 GB for the SAP archive and the PBS indices is required. In total, you have saved 350 GB! In the same way, you can also access productive SAP archives from the test or development system using the PBS indices.

Is there a recommended retention period for the documents in the database?
This is a question of confidence and adaptation of customer-specific programs. Many customers start by archiving completed documents from previous fiscal years and index them using the PBS software. If you have adapted all programs to the archive access, the runtime can also be reduced. There are PBS customers that archive monthly and immediately archive every document that has been completed and thus cannot be changed any more. Most of the documents have a retention period of 3 months, however, to assist end-of-quarter closings only from the database due to the runtime.

How fast is the access to the SAP archive via the PBS indices?
The access to the SAP archive via the PBS indices is at least as fast as the database access. Precondition is that the SAP archive is still in the file system of the operating system. If you use a database server and one or more application servers, the SAP archives and the PBS indices should not be on the database server because it is completely reduced from any load due to the archive access. You can also define a special archive server using the NFS mount. If you archive for the first time and the database has already some performance problems, archiving and PBS indexing have an acceleration effect because the database now works more performantly again.

Is it also possible to have access to SAP archives if they are removed to external storage systems (optical data carriers or content servers)?
The PBS programs recognize if the SAP archive is in the file system, or has been removed via ArchiveLink to an external storage system without further Customizing settings. The data is read automatically from there if required.

What is the data backup concept for the PBS indices?
In the simplest case, the PBS indices can be included like the SAP archives in the normal backup concept for the file system of the operating system. This means that the modified directories are saved after the SAP data archiving and the PBS index setup. The data is not changed until the next archiving run. If damage occurs in the file system disks, you can change the drive at any time and reconstruct the content of the defective disk drive per restore. Consequently, the requirements for the archive server drives are by far not as high as those for the database drives for which the data security is often created by mirroring. Thus the costs for such drives are reduced.

Apart from in CFI and CSD, the PBS indices can also be removed with ArchiveLink to optical archive systems and, nevertheless, be processed using the PBS transactions. However, a copy should be retained in the file system due to the access speed.

How many documents can PBS add to a module for an SAP archiving object?
In theory and in practice any number. If, however, archiving has reached a certain scale and a PBS long-term archiving concept (see this section for more information) has not been realized for the SAP archive, you should contact PBS referring a long-term archiving concept for this SAP archive.

How do you archive PBS indices?
For each PBS module in which the long-term archiving concept for an SAP archiving object has not been realized, a parameter exists in the index setup program. You can define the time from when the data in the PBS index should be transferred using this parameter. If you increase this time in a merge run, all older data is not transferred to the new directory and thus deleted. The disadvantage is that the data cannot be activated again if you need it once more. In this case, the PBS index has to be set up again with the old time. This is not the case for the PBS modules in which a long-term archiving concept has been realized.

What does the term long-term archiving mean in the PBS archive add ons?

The volume of archived documents increases with continuous archiving. This inevitably increases the response times when accessing these archived documents. At the same time, the need to display documents decreases the older they are, and is nearly zero after some years. Without introducing the PBS long-term archiving all indices for an account would always be read even though you want to select specific, mostly younger indices using a filter. The long-term archiving concept now offers the possibility to split the PBS archive so that only exactly defined sections of the archive can be accessed permanently and the other sections are only taken into account according to specific preconditions in the selections. A sensible classification that is recommended by PBS is the classification according to archiving years:
For each new year you start with a new PBS index directory. Data is only added to this new, now current directory. The previous directories remain unchanged from the time a new directory was selected. But you can still access the previous directories according to specific criteria via a PBS table. This way, you can define that only specific users or processes in the background can access older directories. For example, you can assign only the directory of the current year and the previous year to the standard user. This user can then access the data of the current fiscal year and the previous year. Moreover, data from the years before last is available that is still in process or has been closed in the current year or the previous year.

Users with special authorizations can be given access to older directories. If, for example, an auditor of the financial authorities wants to audit a period that was more than four years ago, controlled via a PBS table, you can allow him to access the directories of the last four years (when doing this, you can use the fiscal year authorization in the SAP system to prevent the auditor from being able to select younger periods even though all younger directories and the database are read).

If older directories are no longer required, you can exclude them from access, and remove them to slower media such as optical data carriers or non-online capable media such as backup tapes. Then, if necessary, they are easier to reload and activate for the access. These (reload) processes can be executed using an HSM system.

An archiving concept for the PBS indices is realized with this removal in exactly the same way as for the archiving concept for SAP archives.

Which PBS modules currently have a long-term archive?

At present, the long-term archiving concept is realized in the modules

• CFI for the SAP archiving object FI_DOCUMNT (from SAP R/3 Release 4.0b) and for the SAP archiving object MM_MATBEL (from SAP R/3 Release 4.6c)
• CSD for the SAP archiving objects SD_VBAK, RV_LIKP and SD_VBRK (from 4.0b)
• CCO for the SAP archiving object CO_ITEM (from 3.1i)
• CMM for the SAP archiving objects MM_EKKO (from 4.6b), MM_EBAN (from 4.0b) and MM_REBEL (4.6b)
• CPP for the SAP archiving object PP_ORDER (from 3.1i)
• CPM for the SAP archiving object PM_ORDER (from 3.1i)
• CHR for the SAP archiving object PA_LDOC (from 4.0b)
• CPCA for the SAP archiving object EC_PCA_ITEM (from 4.6c)
• CSL for the SAP archiving object FI_SL_DATA (from 4.6c)
• CWM for the archiving objects RL_TA and RL_TB (from 3.1i)
• CCU for the archiving object CHANGEDOCU (from 4.6c)

 For which other PBS modules or SAP archiving objects is a long-term archive planned?

All other possible long-term archives depend on the customer requirements. This problem does not occur in some modules (CCO, CPP, CPM and CHR) due to the technique that is used (see the corresponding documentation in the administrator manuals of the respective module).

 What should you take into account if a long-term archive is set up in an existing PBS archive?

Normally, it does not change anything for the end user. It is only the administrator who has to add some tasks to the standard procedures:

Additional archive paths (directories) have to be created.
The old archive path must be entered in a PBS table. This table controls which directories are active.
The authorized users have to be assigned to a user group in the respective PBS table for the special access to older directories. Finally, the current archive path and the historical archive paths have to be activated.

Can I adapt our in-house programs to PBS indexing?

There are different methods to adapt your own programs to PBS indexing. First of all, there are logical databases in every SAP module which are mostly used in the financial accounting component of the SAP standard. PBS has its own logical databases, which either replace the SAP standard or define their own. In the former example, it is normally sufficient to replace the name of the logical database with the PBS name in your in-house report. This automatically provides archive access using PBS indexing. In the second example, you program a report as if it were intended to access the database only. The archive is processed by the PBS logical database using the PBS indices.

The next option is the PBS Conversion Tool, which is used when programming with database selects. You can use this tool to copy your own report into a new one. In the copy process, the database selects are converted into a "select into table" and PBS function modules are generated to also retrieve the archive data in this table. Afterwards, the table's data is processed within a loop. This tool is useful in many common applications. More complex reports cannot be converted as easily, as this requires specific information about the data structures in use. PBS provides function modules and external subroutines that enable you to access and retrieve the archived data. Example programs are contained in the "Archive access tutorial" (transaction /PBS/UTIL_ACCESS_ARCH) from Release 4.6C. PBS can assist you in adapting your programs and reports at a reasonable additional cost.

In many cases, especially in the module CFI, PBS already provides enhanced SAP standard reports that automatically evaluate databases and archived data. If you have developed your own reports from SAP standard reports, you can use the PBS solution as a basis for your own modified reports.

 I want to add PBS archive access to a report. The report is based on a logical database.

PBS has its own logical databases, which either replace the SAP standard or define their own. In the former example, it is normally sufficient to replace the name of the logical database with the PBS name in your in-house report. This automatically provides archive access using PBS indexing. In the second example, you program a report as if it were intended to access the database only. The archive is processed by the PBS logical database using the PBS indices.

 I want to use PBS function modules directly to access an archive. Where can I find some examples of this?

Examples of using PBS function modules for accessing archived data are summarized in the Archive Access Tutorial. You can call up the Archive Access Tutorial in SAP Release 4.6C via the transaction /PBS/UTIL_ACCESS_ARCH.