Excuse me, where’s the Global IoT Analytics Unit please?
7 in 10 enterprises worldwide now collecting IoT data
7 in 10 enterprises worldwide now collecting IoT data

Excuse me, where’s the Global IoT Analytics Unit please?

No honestly, it appears some firms actually don’t have a data crunching IoT analytics lab on their premises — but don’t worry, Teradata does.

You know the scene all too well; after all, you’ve been there yourself. The new employee nervously walks into their new office premises on day one and clumsily looks for their workstation location, the water cooler and the perhaps (it’s old school we know, but some people still use staplers) the stationary cupboard.

Those essential functionalities geo-located, the newbie finally plucks up the courage to ask that burning question.

“Umm, excuse me, but could you tell me where the Global IoT Analytics Unit is please?”

Huh? What do you mean? You don’t have a Global IoT Analytics Unit here? How do you actually function and know what’s going on? No honestly, it appears some firms actually don’t have a data crunching IoT lab on their premises.

This revelation would surely come as a great surprise to Teradata. The data analytics company has recently handed round the party hats, sponge cake and fizzy pop to celebrate the creation of a Global IoT Analytics Unit within its Teradata Labs operations in the United States, UK and India.

Cheeky IoT analytics cheese

Cheekily tagged for additional cheesy industry-speak value as a means of deriving value from the so-called Analytics of Things, or AoT (Ed — don’t worry, this latest acronym won’t stick or catch on), these new units will be staffed by ‘special-ops’ teams of data scientists, data engineers and software designers.

The ‘laser focused special-ops tactics’ (no, honest, Teradata said that) in use here will be focused on building new, cloud-based analytic solutions and services to simplify advanced analytics, data movement and database management. All focused on the Internet of Things, obviously.

When he’s not talking about laser focus, coining new industry acronyms and generally saving the planet from unmanaged log data, Teradata Labs president Oliver Ratzesberger is happy to go on the record with a few appropriate platitudes.

“With this announcement, we are making it easier for our customers to move sensor data around, optimise data management systems to deal with the massive volumes of data and run real-time, advanced analytics against streams of IoT data,” said Ratzesberger.

You might like to read: Will cognitive analytics make the IoT smarter?

Data: you’ve got to move it, move it

Actually, those are platitudes or banalities at all; optimising IoT data management systems to cope with the new dataflows emanating from the IoT itself is what it’s all about. Being able to now fundamentally MOVE this data around so that we can apply the right analytics at the right time in the right order could just be why we’ve been banging the rocks together after all, right?

Teradata says it is looking to answer the “why did this happen?” questions that arise in business through using IoT data. Its pre-built analytic functions include new IoT data preparation capabilities and machine learning techniques to understand and detect patterns in machine behaviour.

Additionally, many of the machine learning models generated can be ported to run on virtually any operational environment that can run Java. The Teradata Aster Scoring SDK (software developer’s kit) is supposed to allow data analysts to deploy Aster IoT analytic models into virtually any IoT edge servers, public clouds and in the datacentre.

According to Teradata, “[We are] extending IoT capabilities for Teradata Listener by with connectors that make it easier to acquire and distribute streaming sensor data for analysis. Capturing and managing continuous streams of data is normally complex and labor intensive. These new connectivity options will make it easy and fast for Listener to deliver new data streams of sensor data to the Teradata Unified Data Architecture, either on-premises and in the cloud.”

The real road to data discovery?

The firm insists that its IoT Analytics unit is applying machine learning and advanced analytics techniques to system administration and DevOps tasks. They are applying machine learning to Teradata systems in order to solve complex performance and workload congestion problems in seconds, it says. So do all of Teradata’s moves make sense?

Ian Marsden CTO at Eseye spoke to Internet of Business to say the following, “The real question, is analytics something that can be outsourced and if so is it worthwhile? Our datacentres process many terabytes of data a week, while the external connection would not be able to transmit that much data. As a result, we have no choice but to analyse a certain amount of data internally.”

Sarcastic jibes aside, is Teradata positioning itself well here? Well, many would argue that the need to operationalise data management in this way for data discovery platforms, analytic appliances, enterprise data warehouses and data marts is the real key to the Internet of Things.

Coming next: all new IoT analytics Artificial Intelligence headsets. Okay, just kidding, for now.

You might like to read: 5 reasons why IoT needs data analytics