NEW HORIZONS IN BUSINESS ANALYTICS.

A new star is on the rise in the Data analytics horizon. Big guns who have monopolized the markets and ran the show the way they liked are now trembling before the sheer lightning-speed and performance of SAP HANA.

Short for High-performance Analytics Appliance, HANA combines a lot of what is “future technology” for its competitors, and SAP has done an excellent job pulling it off. In fact the entrance of HANA was so unexpected, even the competitors rolled out by Oracle and the lots are in no way a match for it.

SAP first announced its plans for working on in-memory databases at the SAPPHIRE conference in May 2011, and was immediately met with sharp criticism or rather sarcasm! Oracle CEO Larry Ellison, openly mocked SAP's goals during an event in January.

"Get me the name of their pharmacist," Ellison said at the time. "I mean, I know a lot about in-memory databases. In fact, we have the leading in-memory database, TimesTen. This is nonsense. There is no in-memory technology anywhere near ready to take the place of a relational database. It's a complete fantasy on their part."

In-memory databases store information in a system's main memory, instead of on disk, providing a performance boost. The technology, which Plattner also discussed at last year's Sapphire event, is now beginning to be reflected in SAP BI (business intelligence) products, such as BusinessObjects Explorer.

HANA is available in appliance form on hardware from a number of vendors. Built in part upon existing products at SAP, HANA places data to be processed in RAM, versus reading it off disks, which provides a performance boost.

SAP HANA runs on high-end commodity hardware and provided you have enough memory to run your database, it doesn’t matter what you run it on from a technology standpoint.

Initial certified hardware solutions are still quite expensive – I priced up a 1TB Dell system for $75k – not including disk storage, which will probably double that number at retail price, so think $150k. But really that’s nothing compared to what an equivalent 10TB database cost 10 years ago (SAP HANA compresses 10:1 compared to databases back then).

And worryingly for Oracle, Teradata, IBM and HP, it is nothing compared to what Mainframe or Teradata/Exadata hardware costs.

While HANA so far is positioned primarily for analytic workloads, SAP has long-term designs of porting its flagship Business Suite to the platform, a move that would provide an alternative for SAP customers currently running their transactional systems on databases from Oracle or IBM.

SAP had previously released a Strategic Workforce Planning application for HANA, which allows companies to analyze the effect of changes to their employee base. The new Smart Meter Analytics product will help utilities probe smart meter data for deep insights into customer energy usage patterns, allowing for improved system load forecasting, specialized marketing programs and other benefits, SAP said.

What does this mean for the future state of high-end systems? And in the emerging business world where the Big Data challenge is not just an isolated incident but a ubiquitous condition, is it not reasonable to expect that more and more customers will require more and more of these extremely high-performance machines?

REAL LIFE CASE-STUDYS

What does improved analytics speed mean for business? What is the value added to your business by pulling off super-fast reports and analysis from your databases? What if you could do cost and profitability analyses across millions of records spanning three years and thousands of retail outlets in less than 5 seconds—that is, in real-time—instead of in a few days? How would that change the ways in which you engage with customers, exploit the power of dynamic pricing, squeeze inventory volumes, and accelerate your moves into new markets?

How about a factor of 1,000: what if your systems could give you deep and granular insights into every phase of your business 1,000 times faster than they currently do—how would you exploit all that speed, all that customer-driven foresight, all that potential?

Only a year or so ago, all of these questions would have been largely theoretical simply because traditional IT systems and architectures were limited to delivering only incremental increases in performance, and gains of 25% or 35% were huge.

HANA enhances or rather expands old-school analytics in five dimensions. In the words of SAP executive board member Vishal Sikka, these are:

1)Are we able to ask complex questions? This, Sikka said, means that value is unlocked not merely by crunching through the same numbers more quickly than before, but by “going deeper”: analyzing more records over more time with more variables, and getting answers hundreds or thousands of times faster.
2) Can we ask an unlimited range of questions without having to submit them in advance? Current systems require some preliminary staging of anticipated data sets, which greatly limits the types of questions that can be asked and, more importantly, the types of insights that can be gained. HANA, Sikka said, has no such limitations.
3) Can we interactively arrive at questions? In the old model, a limited set of questions is submitted to an expert who then triggers the analysis in isolation—conversely, Sikka is describing an entirely new dynamic that allows multiple people to look at and question the incoming results on the fly from an unlimited set of perspectives.
4) Are we able to ask questions across data sets that are not only huge, but come in various types: structured and unstructured, month-old and real-time? With absolutely no limitations?
5) And, can people from across the organization—not just BI black-belts—make these interactive inquiries in real time and get real-time answers? Can businesses finally unlock the potential and power within their data that for too long have been trapped in traditional systems that, intentionally or otherwise, give higher priority to internal processes than to customer requirements?

It all depends on which of these dimensions you chose to exploit, and how you go about doing it.

“Our friends at Colgate told me that with their live Hana system, they are now able to run some operational reports 1,000 times faster than before,” said VishalSikka during his keynote speech at SAP’s China SAPPHIRE NOW event in Beijing. The list doesn’t end there.

Nongfu Spring, a rapidly growing bottled-water company based in China whose use of HANA has enabled it to cut its calculation of transportation costs from 24 hours to 3.8 seconds, or more than 15,000x.

Even more astounding is the achievement of is Yodobashi, the electronics retailer in Japan, where 5 million of its 22 million customers are part of the company’s loyalty program. The calculation for rewards and incentives earned by those 5 million loyalty-club members used to take 3 days—but with HANA, it now takes 2 seconds. That’s an improvement of 125,000 times!

As a result, Yodobashi can now engage immediately and more intimately with those 5 million loyalty-club members while they’re still in the store and more likely to buy, which is a massive transformation in how the company can enhance its interactions with its best customers in real time.

“Yodobashi hits 3 of the 5 dimensions we have laid out,” Sikka said, “so just imagine what can happen if a company is able to execute across all five dimensions.”

In short term, we can see IBM posting higher profits from its mainframe sales, HP is in all sorts of organizational troubles and Oracle is focusing on Exadata. But the Mainframe market is set to start tailing off by 2015. Oracle already has an in-memory product called TimesTen in developement, but it is highly doubtful if it will be able to compete with HANA. As for Teradata, they need to realize the big challenge posed before them.

By Sharath with No comments