Dec
01
2020
--

AWS launches new services for its industrial customers

One of the areas that is often left behind when it comes to cloud computing is the industrial sector. That’s because these facilities often have older equipment or proprietary systems that aren’t well suited to the cloud. Amazon wants to change that, and today the company announced a slew of new services at AWS re:Invent aimed at helping the industrial sector understand their equipment and environments better.

For starters, the company announced Amazon Monitron, which is designed to monitor equipment and send signals to the engineering team when the equipment could be breaking down. If industrial companies can know when their equipment is breaking, it allows them to repair on it their own terms, rather than waiting until after it breaks down and having the equipment down at what could be an inopportune time.

As AWS CEO Andy Jassy says, an experienced engineer will know when equipment is breaking down by a certain change in sound or a vibration, but if the machine could tell you even before it got that far, it would be a huge boost to these teams.

“…a lot of companies either don’t have sensors, they’re not modern powerful sensors, or they are not consistent and they don’t know how to take that data from the sensors and send it to the cloud, and they don’t know how to build machine learning models, and our manufacturing companies we work with are asking [us] just solve this [and] build an end-to-end solution. So I’m excited to announce today the launch of Amazon Monotron, which is an end-to-end solution for equipment monitoring,” Jassy said.

The company builds a machine learning model that understands what a normal state looks like, then uses that information to find anomalies and send back information to the team in a mobile app about equipment that needs maintenance now based on the data the model is seeing.

For those companies who may have a more modern system and don’t need the complete package that Monotron offers, Amazon has something for these customers as well. If you have modern sensors, but you don’t have a sophisticated machine learning model, Amazon can ingest this data and apply its machine learning algorithms to find anomalies just as it can with Monotron.

“So we have something for this group of customers as well to announce today, which is the launch of Amazon Lookout for Equipment, which does anomaly detection for industrial machinery,” he said.

In addition, the company announced the Panorama Appliance for companies using cameras at the edge who want to use more sophisticated computer vision, but might not have the most modern equipment to do that. “I’m excited to announce today the launch of the AWS Panorama Appliance which is a new hardware appliance [that allows] organizations to add computer vision to existing on premises smart cameras,” Jassy told AWS re:Invent today.

In addition, it also announced a Panorama SDK to help hardware vendors build smarter cameras based on Panorama.

All of these services are designed to give industrial companies access to sophisticated cloud and machine learning technology at whatever level they may require depending on where they are on the technology journey.

 

May
14
2018
--

Xage introduces fingerprinting to protect industrial IoT devices

As old-school industries like oil and gas increasingly network entities like oil platforms, they become more vulnerable to hacking attacks that were impossible when they were stand-alone. That requires a new approach to security and Xage (pronounced Zage), a security startup that launched last year thinks it has the answer with a concept called ‘fingerprinting’ combined with the blockchain.

“Each individual fingerprint tries to reflect as much information as possible about a device or controller,” Duncan Greatwood, Xage’s CEO explained. They do this by storing configuration data from each device and controller on the network. That includes the hardware type, the software that’s installed on it, the CPU ID, the storage ID and so forth.

If someone were to try to inject malware into one of these controllers, the fingerprint identification would notice a change and shut it down until human technicians could figure out if it’s a legitimate change or not.

Whither blockchain?

You may be wondering where the blockchain comes into this, but imagine a honey pot of these fingerprints were stored in a conventional database. If that database were compromised, it would mean hackers could have access to a company’s entire store of fingerprints, completely neutering that idea. That’s where the blockchain comes in.

Greatwood says it serves multiple purposes to prevent such a scenario from happening. For starters, it takes away that centralized honey pot. It also provides a means of authentication making it impossible to insert a fake fingerprint without explicit permission to do so.

But he says that Xage takes one more precaution unrelated to the blockchain to allow for legitimate updates to the controller. “We have a digital replica (twin) of the system we keep in the cloud, so if someone is changing the software or plans to change it on a device or controller, we will pre-calculate what the new fingerprint will be before we update the controller,” he said. That will allow them to understand when there is a sanctioned update happening and not an external threat agent trying to mimic one.

Checks and balances

In this way they check the validity of every fingerprint and have checks and balances every step of the way. If the updated fingerprint matches the cloud replica, they can be reasonably assured that it’s authentic. If it doesn’t, he says they assume the fingerprint might have been hacked and shut it down for further investigation by the customer.

While this sounds like a complex way of protecting this infrastructure, Greatwood points out that these devices and controllers tend to be fairly simple in terms of their configuration, not like the complexities involved in managing security on a network of workstations with many possible access points for hackers.

The irony here is that these companies are networking their devices to simplify maintenance, but in doing so they have created a new set of issues. “It’s a very interesting problem. They are adopting IoT, so they don’t have to do [so many] truck rolls. They want that network capability, but then the risk of hacking is greater because it only takes one hack to get access to thousands of controllers,” he explained.

In case you are thinking they may be overstating the actual problem of oil rigs and other industrial targets getting hacked, a Department of Homeland Security report released in March suggests that the energy sector has been an area of interest for nation-state hackers in recent years.

Jan
17
2017
--

Alchemist Accelerator closes $2.5 million to back enterprise tech startups

IOTGlobe Alchemist Accelerator has closed a $2.5 million fund for very early-stage investments in tech startups serving large organizations in an era when robotics, artificial intelligence and the Internet of Things are fundamentally changing the way they operate. Limited partners in Alchemists’ second fund included Johnson Controls, Ericsson and Analog Garage, a VC arm of Analog Devices.… Read More

Powered by WordPress | Theme: Aeros 2.0 by TheBuckmaker.com