For industry users, Amazon Web Services (AWS) may still seem like the new kid on the block, but the company has been doing business since 2006. For most of its early years, AWS provided cloud-based computing and storage services to start-up companies that gladly embraced its cashflow-friendly pay-as-you-go model rather than investing a big chunk of their start-up capital in IT equipment.
Now AWS is taking its business model to industrial companies. But why now? Cloud-computing is ubiquitous in nearly every other industry. Banks and insurance companies, online retailers, and even whole governments moved their IT to the cloud years ago. Manufacturers are typically the last companies to adopt new technologies, often for good reasons. Manufacturing assets such as production machinery and process equipment are expensive, but they are built to run for many years and even decades. That means that the window of opportunity for new technologies is rarely open. Also, equipment for a factory or production line is sourced from many different machine builders whose machines perform a single function that is later integrated into a whole line, so openness and connectivity is not built in from day one. Finally, as hackers increasingly target industrial infrastructure, conservative manufacturers loath to expose their equipment to the dangers of a life online and its associated liability.
AWS Transformation Day
AWS recently invited about 1000 members of its community, the press, and industry analysts to its Transformation Day event, held in Munich, Germany. This flashy event included testimonies by several high-profile customers, followed by specific breakout sessions run by both AWS and partner companies. An exhibition with about 40 partners rounded off the event. Several industry-related companies exhibited, including Siemens, the event's platinum sponsor, as well as Autodesk, Beckhoff, Cadence, Foghorn, Gräbert, Infor and Tulip.
While Siemens has hosting agreements with other platform suppliers (SAP HANA, Microsoft Azure), the most visible relationship is with AWS. Siemens MindSphere, the company’s “open IoT operating system”, can be hosted locally on a user’s own server, but most companies will likely opt for cloud-hosting services from providers like AWS. In addition to the high level of security and availability offered by such platforms, users can also take advantage of additional analytics tools available on the platform. Most will also find the pay-as-you-go model more attractive than buying their own equipment and hiring personnel to manage it.
Besides platforms like MindSphere, users can elect to have other industry software hosted in the cloud. For which applications does it make sense? Any non-time-critical, transactional software like visualization or MES or asset management software is suitable. While complex PLM software is still best run locally, the source file can be located in the cloud and updated from other device types like tablets, albeit with less functionality (view only, or small edits).
Everything in the cloud?
For industry users, AWS offers a cloud-based environment with a suite of sophisticated applications and tools. AWS Greengrass lets users run local compute, messaging, data caching, sync, and ML inference capabilities for connected plant devices. AWS Lambda hosts user code on a pay-by-run basis, meaning that the user is charged only when code is running. According to AWS, the combination of Lambda and Greengrass ensures that IoT devices can respond quickly to local events, interact with local resources, operate with intermittent connections, stay updated with over the air updates, and minimize the cost of transmitting IoT data to the cloud. ML Inference allows users to perform machine learning inference locally on Greengrass Core devices using models that are built and trained in the cloud.
AWS Greengrass extends AWS to devices, so they can act locally on the data they generate, while still using the cloud for management, analytics, and durable storage. Greengrass allows users to use familiar languages and programming models to create and test device software in the cloud, and then deploy it to devices. It authenticates and encrypts device data at all points of connection using security and access management capabilities of AWS IoT Core.
Whose app is it anyway?
Plant and automation engineers work with software all the time, but their focus is on the process and the machine, and not on the IT infrastructure that supports a growing amount of software in the factory. Modern, complex IT infrastructures require skills to deploy, maintain, and optimize the environment in which software runs. For this reason, software today should be maintained by specialists (read: IT admins) rather than by plant and automation engineers.
The message at Transformation Day was that AWS has expanded its focus to industry software, and is partnering with a growing field of providers, from established automation suppliers to start-up-like industry analytics specialists. AWS made its presence known in the industrial world with a first-ever appearance at the Hanover Fair in 2017 and was also present in 2018. As it grows its footprint in the industrial world, AWS already claims a whopping $24 billion in revenue with recent YoY growth of 49%. With the promise of real value in hosting industry software, part of that growth in the near future likely will be industrial companies as they learn to deploy and manage their software more efficiently.