G.E. Plans App Store for Gears of Industry.
NEW YORK, Aug 05, 2015 (BUSINESS WIRE) — As a digital industrial company focused on answering the unique needs and scale of customers across aviation, energy, healthcare and transportation, GE has announced plans to enter the cloud services market with Predix Cloud.
G.E. is announcing on Wednesday a push into computer-based services, connecting sensors that are on machines to distant computing centers where data will be scanned for insights around things like performance, maintenance and supplies.GE, a company that sells consumer appliances as well as jet engines and other complex industrial equipment, today announced Predix Cloud, a version of its Predix machine data analytics software that will be accessible to companies as a cloud service.It is important to note that although financial details of the development in providing Infrastructure as a Service (IaaS) are yet to be disclosed, but industry experts are seen to hold confidence in GE to dominate the segment. Last month Chrysler recalled 1.4 million vehicles after WIRED reported a security vulnerability that allowed hackers to take remote control of someone else’s Jeep.
The world’s first and only cloud solution designed specifically for industrial data and analytics, this platform-as-a-service (PaaS) will capture and analyze the unique volume, velocity and variety of machine data within a highly secure, industrial-strength cloud environment. The company plans to spend about $500 million annually building the business, according to the executive in charge. “We think it will change the industrial world,” said William Ruh, the head of G.E.’s software business. “We’re talking about where an industrial company goes to get its applications.” The move highlights how important the so-called Internet of Things, a term for matching sensors with cloud-computing systems, has become for some of the world’s biggest companies. GE businesses will start to migrate their software to the Predix cloud service in the fourth quarter of this year, and it will become available to customers next year, GE said in a statement on the news today. GE has launched a cloud service based on its industrial equipment monitoring and analytics platform, Predix, which it is aiming at industrial IoT applications in most of the verticals it already serves – aviation, manufacturing, automotive, energy and healthcare. Predix Cloud will drive the next phase of growth for the Industrial Internet and enable developers to rapidly create, deploy and manage applications and services for industry.
The service represents another attempt on the part of GE to get into the business of analyzing and acting on data from Internet-connected industrial devices. With $4B in software revenues in 2014 and projected software revenues of $6B in 2015, GE continues to grow its investment in software. “Cloud computing has enabled incredible innovation across the consumer world. GE officials told Reuters that Predix Cloud is designed to manage, analyze and store machine data in real-time, using what it calls a “gated community model” that ensures cloud tenants deserve there place in the industrial ecosystem. But as frightening as the Jeep hack was, a bigger danger for IoT customers may lie in the clouds that host all the data these devices collect, whether cars, thermostats or fitness trackers.
Your devices send you feedback but also sned it up to a cloud where that data can be parsed and analyzed and perhaps sent back to you in the form of your weekly fitness updates or energy savings. IBM bought Softlayer in 2013, then launched the Bluemix development platform the following year with the hope that developers will create applications on Bluemix and run them on Softlayer. GE, which generated more than $148 billion in revenue in 2014, said in October that it was expecting to bring in $1 billion in “industrial Internet” revenue annually. This should ensure better visibility for detecting and isolating any threats, and preventing any data loss, the company said. “A cloud built exclusively to capture and analyse machine data will make unforeseen problems and missed opportunities increasingly a complication of the past”, said Harel Kodesh, vice president, general manager of Predix at GE Software. Unlike Amazon’s AWS service, which is open to everyone from large companies to individual users, GE is specifically targeting the industrial market—think connected jet engines, medical equipment, and mail-sorting machines.
The goal is to have a place where manufacturers can buy the software applications they need, something like the way consumers pick up a new game in the mobile app stores of Apple and Google. GE says Predix Cloud should play nicely with other cloud fabrics – for example, it can tap into Pivotal’s Cloud Foundry PaaS to help develop and deploy applications. As you might expect, Kodesh says one of the main selling points for going with the GE infrastructure offering, is that it plays nicely with the software development part of the platform.
While the company has undeniable insights into how the industrial world works, its efforts in computing pale in comparison with IBM, which is increasingly targeting fields like health care and industrial operations, long the domain of G.E. The pitch is that GE is (or will be) first to market with cloud computing built especially for these jobs, not consumer and general-purpose IT applications Kodesh said. This apart, Predix Cloud provides advanced connectivity-as-a-service for these industrial assets to enable provisioning of sensors, gateways and software-defined machines. IBM has hired experts from various industries to sharpen the algorithms of Watson — the IBM computer famous for winning on the television game show “Jeopardy!” — to meet the needs of, say, an oncologist or a supply chain manager. “In the past we were a pure data-processing company, we didn’t want to own the data or build the software applications,” said John Kelly, who heads IBM’s Watson and analytics businesses. “We’ve crossed a line we wouldn’t cross before.” Microsoft is also hoping to persuade companies to use its cloud computing, called Azure, for storing and searching industrial data. More recently the company gave itself another shot in the arm with last year’s acquisition of API Healthcare, a provider of real-time analytics services and employee management solutions for the healthcare industry.
Even Google, which makes almost all of its money from categorizing web pages and putting ads on them, is positioning itself as a data-analysis company for business. That raises the stakes for G.E. “It’s a whole new competition for them,” said Yefim Natis, a senior analyst with Gartner. “To run businesses in a modern way you have to be analytic and predictive. Kodesh said that the Predix platform generated $4 billion in software revenue last year, and the company sees additional revenue opportunities with the infrastructure offering. General Electric calls its new product Predix Cloud, the latest addition to of a set of Predix developer tools, originally designed to work only with with GE equipment but later opened up to all connected gear.
He was careful not to criticize other IaaS vendors, but simply pointed out that they are typically designed for consumers or for general IT requirements. The service is not just a place to run applications, but a platform that includes tools for quickly and easily building applications that integrate with IoT devices. The beauty of getting all that streamed data in near-real-time, is that anomalies can be flagged right away, and service can be performed before there is a catastrophic failure. There is the PaaS that can run atop Amazon Web Services for development and testing, but actual deployment will happen in Predix data centers and infrastructure, he said. “This is completely different from all the consumer and IT clouds,” he noted.
GE, which makes everything from jet engines to CAT scans to power generation gear, knows a lot about machinery and the data it throws off and it knows that secure connectivity to those machines and data security is paramount. “We want to make sure no one can get into that power plant,” he noted. Unfortunately, details are sparse as to exactly how the company will keep data protected, beyond following standard practices such as encrypting data as it flows from a device to the cloud. Three years ago when GE started talking up this industrial internet opportunity, it estimated that connecting these big machines could boost global GDP by $10 trillion to $15 trillion in 20 years. But Harel Kodesh, CTO of GE Software, did explain that the company is using software defined networking to isolate each “layer” of an application and observe and limit what data each layer can access. By allowing just anyone to upload an application to its servers, companies like Amazon and Google in theory risk opening themselves up to attacks where one customer’s application breaks free of its virtual environment and steals data from another customer’s app.
This type of attack is mostly hypothetical thus far, but by carefully screening who can use its service, GE can reduce its chances of falling victim to such a scheme. The catch, Kodesh says, is that all these extra security layers and custom features require more processing power, which will increase the cost of using its cloud. “But the industrial market is so concerned about security that we think it’s the right service to offer them,” he says.
As scary as a jet engine sending data to the Internet may sound, there’s a huge value in capturing more information from all of the machines that we use. As such, businesses will be able to take advantage of its optimized security and data structure offerings while maintaining and interoperating within existing solutions. In its labs and factories and on the ground with customers, GE is inventing the next industrial era to move, power, build and cure the world. www.ge.com