SAP flings out one-hub-to-rule-and-crunch all the data. Yes, a hub
No, the data isn't centralised... What do you mean confusing?
Enterprise giant SAP is taking on silos with its latest offering that aims to centralise data processing and governance - but not storage.
SAP’s Data Hub, launched on Monday at an event in New York, looks to capitalise on the vast amounts of data companies collect, especially when they are from an increasing variety of sources.
The firm claims the product will find the data a company needs, and manage it centrally, without having to move all their data into one place before that can happen.
Speaking at the launch event, SAP CEO Bill McDermott said that complexity would “kill companies fast” - instead they want “simple”, scalable and efficient way to manage their data.
However, according to a study carried out by SAP and published to coincide with the launch, some 74 per cent of IT decision-makers told the firm their data landscape was so "complex" it limited agility.
Greg McStravick, SAP president of database and data management, emphasised the problems badly managed data can bring, saying that although data offers huge potential, if firms aren’t careful, data “can become the corporate cholesterol for your enterprise”.
He said that there were three main challenges for big business: data governance, refining the data and sharing the data across these systems.
The SAP Data Hub* aims to manage these elements in one place but, crucially, does not require the data to be in one place.
“You’re moving from a world of centralising data in one location - ETL data, move it into a central repository - to a world of centralising the data management,” McStravick said.
“It’s the movement, orchestration, monitoring and governance of the data… you leave the data where it resides.”
The Data Hub will bring in SAP and non-SAP data, and is designed to work with on-premise, cloud and hybrid cloud.
The governance aspect aims to give customers a better and more detailed view of all the data they use, including who has accessed or changed it, wherever it resides.
SAP said that the data would be brought into the system via a powerful data pipeline, with information from across the range of sources harmonised, transformed and processed.
The hub will provide “fast execution of pipeline activities by distributing computational tasks to native environments where the data resides”, it said, while a highly graphical environment will manage and create data flows in real-time.
The execution can be pushed down to other platforms, and SAP said customers would be able to use processing capabilities in solutions like SAP HANA, Apache Hadoop and SAP Vora.
In a demo of the solution, software architect Franz Faerber, said that Data Hub “isn’t just a collection of technologies… it’s integrating into one consistent system - governance, pipelining and data sharing in one shot”. ?
*McStravick did acknowledge that describing something that does the opposite of bringing data together in one place a hub might be confusing. He said they “hesitated” to use the name as it could have implied the data was centralised. Evidently it didn't stop them - perhaps their imaginations ran dry.