Choosing a new NLS solution

As Informatica recently announced that they are discontinuing support for their ILM Nearline storage, many companies are looking for a suitable replacement. There are a lot of options to choose from – be it the in-house SAP IQ, traditional databases such as Oracle, MSSQL or DB2, Big Data technologies like Hadoop (Spark, Impala or HIVE) or even leveraging Cloud solutions such as Microsoft Azure or AWS for nearline data. From our consulting experience, we recommend going with Hadoop – a highly scalable yet cheap to run platform for data storage and data processing. (You can read more on choosing the best option in the article by Goetz Lessmann – Benchmarking of SAP BW NLS solutions)

Moving archived data is not as easy as it seems

But apart from selecting the right technology that could replace Informatica, there is yet another challenge – moving the data. It’s not that bad if you are running a small system and you are not yet on HANA. Then you can easily move your data back to SAP BW, and then to the new NLS solution. The trouble starts with high data volumes, and especially with SAP HANA systems – often the size of HANA or lack of available database space rules out the option of transferring the archived data back to the online database. And in most cases, buying a bigger HANA box is just too expensive.

When you turn to SAP notes, it seems that there is no easy way out, unless the Finance department decides to give you a late birthday present. But if this is not your case, there is still the option to reach out to a service and software provider specializing in data management.

Benefits of a direct data migration between two NLS solutions (and how we did it)

Recently we’ve been approached by a multinational oil and gas company with a request to migrate 270TB of archived data to a new NLS storage – without reloading it back to SAP BW. So, we couldn’t help but to tinker a bit with our proven OutBoard DataTiering solution and come back with software which can seamlessly move data between different NLS storages, in near zero downtime. The whole project was delivered within the set deadlines and it did not affect the customer’s reporting capabilities.

The benefits of migrating data directly between two NLS solutions include:

  • Low effort: Migration and monitoring cockpit for significant automation.
  • High speed: Migration is much faster than a reload & archive. Data is not reloaded back to SAP BW InfoProviders. No need to create new DAP’s. Automation of processing. No downtime for reporting. Parallelization to support multiple background jobs processing.
  • No memory storage issues: Reload of NLS data back to SAP is not needed. Migration happens old NLS -> new NLS directly.
  • Parallelization: NLS data can be migrated in one “big-bang” or sequentially InfoProvider-based, even request-based. Processing using multiple background jobs.
  • All storages supported: Migration between different NLS vendors such as Informatica, PBS, SAP IQ and Datavard OutBoard. Supported storages of new and old NLS such as SAP IQ, Hadoop, MS SQL, DB2, ORACLE, file, cloud, AWS, Azure.

This is good news for anyone considering a change of their current NLS storage.