IoT at the Edge
Showing results for 
Search instead for 
Did you mean: 

Hybrid Internet of Things—helping industries prepare for the future with data insights




Peter Moser.jpgBy Peter Moser
Director and Account Chief Technologist,
HPE Presales Global Accounts

Good ideas are just that—ideas. Then someone with a real business opportunity translates them into something actionable, through innovation, to create a platform or solution that changes the game. (For example, the Apple iPhone platform and its broad ecosystem changed the cell phone industry.) That raises the competitive bar enough to force the competition and others to either follow or falter.

That’s what’s happening with the hybrid Industrial Internet of things (IIoT). Predictive analytics is the desired end game for many companies that want to take their insights and success to the next level. It is foresight that will enable the next industrial revolution.

For this discussion let’s use oil and gas (O&G) to illustrate what’s driving businesses to hybrid IIoT and what they expect to get from it in return. The price per barrel of oil[i] is a key driver behind much of the change going on in O&G. Note the volatility. Since it is a commodity there isn’t much an individual company can do to impact the selling price.

Price of a barrel of oil (USD).pngIn 2008 the price of oil was more than $150 USD per barrel. The business focus was drill, drill, drill to maximize revenue (outcome). Fast forward to 2016 when oil dropped below $30 per barrel. The focus shifted to optimize, optimize, optimize to reduce costs (outcome).

In today’s volatile oil market, sustained optimization isn’t enough. Companies also want the agility to respond quickly when the market changes and the elasticity to expand and contract more easily. They are looking to collaboration and predictive analytics to achieve these goals. This can be a challenge when most of the data is still trapped in silos; operations technology (OT) and information technology (IT) teams still operate mostly independent of one another; and the traditional technologies in use are not capable of handling the demands necessary to achieve these outcomes consistently.

So what are some O&G companies doing in response?

  1. Many are making organizational changes to bring IT and OT closer together so the two better collaborate and accelerate innovation and outcomes.
  2. They are deploying more sensors and doing more analytics at the edge where a large volume of valuable data is being generated but is underutilized.[ii]
  3. They are using analytics, coupled with deep learning, to achieve predictive insight to enhance operational efficiencies and profitability and to better predict changes in demand—foresight versus hindsight.

The O&G industry has used analytics for decades so what’s different now?

Oil-Gas.pngFirst, the price of sensor technology is rapidly declining, so it is economically feasible to deploy more sensors to collect more and different data to create better quality and a broader range and variety of insights. Like most analytics, more data with better quality produces fewer false positives.

Now that more data is being created, how do you manage it to get the insights you need? Today, most of the data is sent to regional or primary data centers for processing, because most remote sites lack IT staff, data scientists, and other subject matter experts. And the environment isn’t conducive to traditional IT assets like racks, servers, storage, and networking.

To transport all the data to a regional or primary data center (or the cloud), network connectivity options are VSAT (satellite), 3GPP (cellular), radio, and fiber. Most are either unreliable, low bandwidth, or costly. The availability and quality of each varies by country and region which further compounds the problem.

You could continue to send all of the data to a regional or primary data center for processing but the laws of physics still apply regarding distance and bandwidth, not to mention cost and reliability. Additionally, most of the data is telling them “everything is okay.” So why send it over the network in the first place?

So what do you do? Process the data at the point of creation and ingest. If some of those insights involve a gas pocket the driller is about to hit, waiting 10 to 15 minutes for a signal from a regional data center isn’t a good option. Milliseconds or seconds would be preferred. Or if you’re a driller trying to stay in the pay-zone geology. Minutes matter!

So how do you overcome these challenges to achieve the overall, sustained, optimization you are striving far, with the agility to react with the market?

Answer: Hybrid IIoT.

Integrated_systems.pngNow, remote locations like oil rigs and platforms, chemical plants, and pipelines can use innovation in edge computing, communications, and cloud to ingest, analyze, and create insights at the location versus sending it synchronously to a remote data center. With the right enabling software stack on top, the data can be better utilized by more than one application or user and secured for its entire life.

So now you don’t have to upgrade the network, because you’ve solved the latency challenge. You can look at ALL the data, which improves the quality and variety of insights that you can generate. You can secure the data and insight immediately. And the data never has to leave the country—important because a growing number of sovereign nations are passing legislation that prevents data from leaving their country.

We designed an edge platform to overcome some of the limitations found with traditional IT assets like rack-mount devices. It’s suitable for the conditions found at most remote locations, which vary greatly. It can be mounted in a closet, under a desk, on a wall, or in a rack. It's ruggedized to handle a wider range of temperatures than rack mount devices. It is cartridge based so it has the processor, memory, storage, and networking on the card. If you have a failure or need to add capacity, a non-IT person can plug one in.

Edge computing lets you apply what you save on an expensive network upgrade to getting more value and better outcomes from the data you have. If you still need to send the data to the data center, then you can do it asynchronously or via some other more cost effective means.

The takeaway

Innovation.pngHybrid IIoT helps usher in the next industrial revolution by giving industry access to more data that you can actually analyze, create insights in milliseconds or seconds, do it securely, without a heavy dependence on a wide area network or a local IT configuration or staff.  It changes hindsight to foresight.

If you’d like to learn more, check out these resources:

ESG whitepaper: HPE Solutions for Internet of Things Analytics

HPE blog: How to get started with IoT? Start small, think BIG

HPE Blog: Operational Technology, the Often Overlooked Technology in the Industrial IoT

HPE blog: 5 challenges of Industrial IoT: Edge computing to the rescue

HPE blog: 6 lessons learned in Industrial IoT





Peter Moser.jpgAbout Peter Moser

Peter Moser leads a team of chief technologists and enterprise architects who provide innovative and strategic solutions to HPE’s most strategic global customers for cloud, mobility, security, and analytics. He has been with HPE for 20 years and has held numerous technical and leadership roles in IT, solution centers, consulting, and technical sales.


Hybrid IoT - Helping industries.png

Empowering the Digital Enterprise to be more efficient and innovative through data-driven insights from the Internet of Things (IoT)
About the Author