Yesterday’s launch of SAP Business Suite on HANA comes almost exactly 20 years after the debut of SAP’s game-changing R/3 ERP software.
At the launch, chairman and co-founder of SAP Hasso Plattner described how the team came up with HANA, and why he believes it will take SAP forward into the next era of IT.
“Fourteen years after the launch of R/3 – six years ago – I asked team to look at a new design and new features for our software,” he says. “We proposed to re-invent enterprise systems, redefine them with all the knowledge we have now and figure out what an ideal system would look like.”
The wish-list of features to be included in the hypothetical new system included the following:
* All active data should be in memory;
* It would leverage massively parallel computers – even though multiple-core hardware wasn’t yet available, systems of up to 60 cores are operational now and will only improve;
* It had to use a design thinking methodology;
* The new system would simplify data models – by radically reducing the system on paper, overhead could be reduced to about one-tenth;
* It would bring OLTP and OLAP together – although these traditionally handled separately, Plattner believes this thinking is incorrect;
* A new system would enable instant business intelligence on transactional data – rather than the hours-long batch-like systems currently employed;
* Users would be able to have live conversations around data, with instant information availability;
* The system would offer a response of less than one second, for all activities;
* It would include a virtual data warehouse for multiple data sources and types – the data warehouse would no longer be simply a reporting tool for OLTP but would consolidate data from multiple resources, data types;
* There would be no aggregate or materialised cubes – OLTP reporting would go back into the transactional system to produce data entries that are well-organised and dynamic;
* There would be views on views, allowing business to create information out of data and change it on the fly without the limitation of pre-fabricated models that don’t allow that all companies are different;
* The aggressive use of maths would allow for accurate and instantaneous planning and forecasting; and
* The system would include interactive analysis and planning.
“These were the features we defined at the start of the project six years ago,” Plattner says. “We didn’t think about enabling the mobile workforce then but added it later and have enabled the system to support it.”
Within three years, Plattner says, the design team could report to SAP that the envisioned project was possible and that the company should commence a project to build such a database – and so the HANA Project was launched.
“The goal was to design and build a new in-memory database, and two years ago we were able to start proof-of-concept projects with our customers.”
SAP has come under fire over the last couple of years for developing a database management system in the face of strong competition from existing market leaders Oracle and IBM.
“Three years ago, it was said that SAP would never be able to rewrite 400-million lines of code to take advantage a new in-memory database,” says Plattner. “My spontaneous reaction was: if SAP can’t do it, no-one can.”
The HANA Project was developed by a distributed development team working all over the world, and collaborating with one another over the Internet and via video conferencing.
“Today we are launching what we proposed three years ago: a non-disruptive change to Business Suite, on HANA, with unbelievable performance results,” Plattner says.
“Many didn’t believe it was possible, but we have shown that it is.”
At the launch, chairman and co-founder of SAP Hasso Plattner described how the team came up with HANA, and why he believes it will take SAP forward into the next era of IT.
“Fourteen years after the launch of R/3 – six years ago – I asked team to look at a new design and new features for our software,” he says. “We proposed to re-invent enterprise systems, redefine them with all the knowledge we have now and figure out what an ideal system would look like.”
The wish-list of features to be included in the hypothetical new system included the following:
* All active data should be in memory;
* It would leverage massively parallel computers – even though multiple-core hardware wasn’t yet available, systems of up to 60 cores are operational now and will only improve;
* It had to use a design thinking methodology;
* The new system would simplify data models – by radically reducing the system on paper, overhead could be reduced to about one-tenth;
* It would bring OLTP and OLAP together – although these traditionally handled separately, Plattner believes this thinking is incorrect;
* A new system would enable instant business intelligence on transactional data – rather than the hours-long batch-like systems currently employed;
* Users would be able to have live conversations around data, with instant information availability;
* The system would offer a response of less than one second, for all activities;
* It would include a virtual data warehouse for multiple data sources and types – the data warehouse would no longer be simply a reporting tool for OLTP but would consolidate data from multiple resources, data types;
* There would be no aggregate or materialised cubes – OLTP reporting would go back into the transactional system to produce data entries that are well-organised and dynamic;
* There would be views on views, allowing business to create information out of data and change it on the fly without the limitation of pre-fabricated models that don’t allow that all companies are different;
* The aggressive use of maths would allow for accurate and instantaneous planning and forecasting; and
* The system would include interactive analysis and planning.
“These were the features we defined at the start of the project six years ago,” Plattner says. “We didn’t think about enabling the mobile workforce then but added it later and have enabled the system to support it.”
Within three years, Plattner says, the design team could report to SAP that the envisioned project was possible and that the company should commence a project to build such a database – and so the HANA Project was launched.
“The goal was to design and build a new in-memory database, and two years ago we were able to start proof-of-concept projects with our customers.”
SAP has come under fire over the last couple of years for developing a database management system in the face of strong competition from existing market leaders Oracle and IBM.
“Three years ago, it was said that SAP would never be able to rewrite 400-million lines of code to take advantage a new in-memory database,” says Plattner. “My spontaneous reaction was: if SAP can’t do it, no-one can.”
The HANA Project was developed by a distributed development team working all over the world, and collaborating with one another over the Internet and via video conferencing.
“Today we are launching what we proposed three years ago: a non-disruptive change to Business Suite, on HANA, with unbelievable performance results,” Plattner says.
“Many didn’t believe it was possible, but we have shown that it is.”