Dieting for Databases

Feature Article | September 5, 2007 by admin

Clear division of labor within a group: As a sales company of HEAG Südhessische Energie AG (HSE) and Stadtwerke Mainz AG, ENTEGA Vertrieb GmbH & Co. KG supplies almost a million people in the Rhine-Main region of Germany with power and water. The workforce of 200 at the subsidiary ENTEGA Service GmbH in Darmstadt handle accounting and receivables management and are responsible for all meter services such as reading, installing, and dismantling. The company is also tasked with the upkeep of all IT systems at ENTEGA.
ENTEGA Service uses the SAP solution IS-U (known today as SAP for Utilities) to process customers’ contractual and billing data. When compiling reports, data is transferred to the business warehouse function of SAP NetWeaver Business Intelligence (SAP NetWeaver BI) where it is then processed. However, due to the large volume of billing data that energy suppliers need to process, data volumes in SAP NetWeaver BI grew dramatically – particularly InfoCubes and the database tables of the Persistent Staging Area (PSA). Billing documents are stored in the PSA, which quickly leads to the accumulation of huge amounts of data as several hundred data records are created for each document. The data volumes can grow by up to 20 GB per month.

Time wasted for queries and backups

“SAP NetWeaver BI was already quite ‘full’. We were in urgent need of a solution,” recalls Jens Schwarz, IT specialist at ENTEGA. The situation was further compounded because billing and data exchange processes were being changed over to the multi-client model, which generates double the amount of data because network accounts are created for both the supplier and customer.
This flow of data generated significant and costly outlay for data storage and also resulted in administration problems for ENTEGA Service. As a result, employees had to constantly monitor storage volumes to avoid bottlenecks during simulation runs for balance sheet cutoffs, since these runs are used to project consumption for the calendar year. The company also struggled with long run times for queries and backups. This was particularly problematic in view of the limited capacity for backups, which meant the runs had to be carried out within a specific timeframe.

Ensuring access to archived information

The archiving application of SAP was initially discussed as a possible solution. However, once data has been archived, it cannot be accessed directly with this application. Instead, the data first needs to be loaded back into the business warehouse, which is an extremely time-consuming process. This made the solution unsuitable for ENTEGA’s needs because, as an energy provider, the company is legally required to be able to access data for a period of ten years in case an official audit is carried out. This applies in particular to data that relates to annual reports, sales and inventory statistics and balance sheet cutoffs.
“Our only options were either to keep everything available online and set up a sufficiently large database for this or to introduce a nearline solution. We opted for the latter,” explains Schwarz. The solution enables the company to delete data that is not needed on a daily basis from the relational online database and store it in the nearline memory instead.

Compressed data saves storage space

The company chose the SAP-certified nearline SAND/DNA Access solution from Hamburg-based supplier SAND Technology. The main appeal of this solution was its ability to greatly compress data while still allowing users to access and analyze it, or to use it as a basis for new ODS objects or InfoCubes. Another benefit of the system is how easy it is to implement the storage solution. The nearline database was up and running for SAP NetWeaver BI by January 2007 – just three months after the introduction phase began. Approximately 30 employees are currently using SAND/DNA Access. They store data in the nearline database as required – for example before balance sheet cutoff runs.
If an audit is announced, users can easily access the compressed data, which is a major advantage for the company, as Schwarz explains: “Once data has been archived in the SAP archive it takes a great deal of time to reload it. Appropriate storage space also needs to be reserved in the online database. However, if the data is compressed but is still accessible, more storage space is available over the long term. And if the online database starts to fill up, we can use the solution to compress data and free up more storage space.”

Lower costs, better performance

Further benefits of the system include the large cost savings it can bring. Because less hard disk space is required, backups take less time. The nearline solution can be used to keep disk space almost constant. This leaner database also helps to keep administration outlay much lower.
Performance is also improved, as Schwarz points out: “Reporting times are much better thanks to the smaller size of InfoCubes. We currently store data for two years in the InfoCubes and move data for the preceding year into the nearline database, deleting it from the InfoCube. As a result, weekly tasks such as creating and deleting indexes also take far less time as everything runs far quicker with such a small amount of data.”
However the main advantage is that ENTEGA Service can now use its available storage space far more efficiently. “We start off with 100 percent of data and end up with some five to eight percent.” This primarily benefits disk space and run times for backups, but the company also needs to procure new hardware less often.

Roland Markowski

Roland Markowski

Tags: , , , ,

Leave a Reply