Richard Buckle
April 4, 2016 · 4 minute read

There continue to be ongoing discussions across the NonStop community about the benefits, real or otherwise, from tapping into data stream analytics. This dates back to the early days of NonStop systems when every processor cycle was important – when it comes to mission critical applications, latency is a very big concern. Very early on there were even concerns about the possible degradation simple application monitoring would produce, so much so that solutions aimed at better visualization of the systems and the networks they supported were late to come to NonStop. Transactions were lean and users could ill afford to enrich them for any reason at all.

However, this is no longer an issue for enterprises moving to the latest NonStop X systems. Based on the Intel x86 architecture, early results are indicating anywhere from 50% to as much as 200% performance gains over those equivalent NonStop i systems based on Itanium processors. Concerns over latency and degraded performance have been replaced with concerns about the amount of available headroom each processor now possesses. For NonStop users deploying NonStop systems with multiple processors oftentimes as many as ten, twelve, or sixteen processors per system, in order to ensure very rapid take-over times in situations where imminent software and / or hardware failures are detected by the OS, even reducing the number of processors in a NonStop X still gives them ample room to process richer transactions.

However today, according to Meg Whitman, HPE CEO, “IT strategy and business strategy are no longer separate, they have become inseparable … every business is a technology business today.” In other words, for business to survive and indeed thrive it needs to exploit technology advances on every occasion and perhaps the most important aspect of such advances is the benefits from being able to predict likely outcomes of customer situations. Knowing, with a degree of certainty that borders on the magical, that customers in a certain demographic will always pick the red sweaters over blue ones brings with it enormous benefits felt all the way from the cashier to the supply chain managers.

Magic is the first thing than came to mind after reading an article To Catch a Thief in the March 15, 2016, issue of Fortune Magazine. “A new generation of tools beyond fingerprints and iris scans can measure qualities like body temperature and blood circulation at a short distance and without alerting the subject,” notes reporter, Jeff John Roberts. Quoting Barnabas Szilagyi, a principal at Capco, a company that helps large banks identify security threats, “Each heart rate is unique. The new tech can sense what’s in your veins, your blood pressure and body heat, and identify with great accuracy who you are.”

More importantly,” Sensors are critical to these biometric tools, but it’s the analytics software that makes them shine.” Citing just one example, Fortune then described an East Coast fraud case where “Capco’s tech first determined that a spike in suspicious bank account applications occurred around the same time every month. It then cross-matched those dates with local shipping records to conclude that the ‘customers’ were linked to the monthly arrival of a ship from a small European country. Such records are just one of hundreds of external data sets that banks can use to spot patterns. Others include illegal ATM transactions and credit card payments to massage parlors—both of which soared when the phony bank customers came to town.”

Clearly, to echo Whitman’s observation, technology is the business as without the technology, separating the good from the bad let alone recognizing the need for red sweaters is only feasible when integrating technology with current business logic. At some point, crucial analysis outcomes need to be presented to transactions in flight and the work involved, even for the most jaded of NonStop pragmatists, will be completely acceptable. That is, almost undetectable by those heavy handed application monitoring products that they were so fearful of not all that long ago.

“It’s the analytics software that makes them shine,” a message that shouldn’t be lost on anyone in the NonStop community and products now coming to market, such as Striim, bring with them the technology the businesses need in order to shine, and shine businesses have to do simply to stay in business! Keys to the analytics software we have today are the algorithms, the modelling and the ease with which they can be tuned to look for patterns we value. Big Data may be all about data lakes but it’s the data streams feeding the lakes that are so important to the mission critical applications running on NonStop systems.

HPE has invested heavily on NonStop and with NonStop X is placing huge bets that shared nothing fault tolerant systems, with permanent availability, will still be driving customer-facing applications where money changes hands. That’s been the sweet spot for NonStop systems for more than four decades and financial institutions continue to depend upon them in this role. But monitoring sensors, picking out the bad guys, and watching for red sweaters to fly off the shelf require a little more than the basics and it is with the arrival of Striim whereby mission critical applications on NonStop can really lift their game.

Our blood may be boiling when we think of the possible intrusions into our lives this is making but seriously? Shouldn’t we be pleased to see our true needs being better addressed as a result and shouldn’t we be pleased to know that, as unique as we most definitely are, we can still open bank accounts, take out loans and obtain cash without becoming suspect of anything unsavory! And shouldn’t we be just as pleased to see our NonStop systems never missing a beat in the process.