Citrix reveals 'USB redirection to cloud' plans to vaporise desktop peripherals Local peripherals driving hosted apps and desktops, for artists and security purposes Citrix is working on something it calls “ USB redirection to cloud” that will allow input devices on local devices to work on virtual desktops. Over the last year, everyone in the remote/virtual/as-a-service desktop business has been talking up their ability to virtualise the graphics-heavy apps that some users wield. It's important for desktop virtualisers to be able to handle such applications because it turns out that demanding apps are in demand in out-of-the-way places where operating a fleet of PCs is hard to do. Oil rigs and mines are often mentioned to The Reg's virtualisation desk as hotbeds of CAD users who can best be served with virtual desktops. Citrix now thinks such users also want to be able to use peripherals like drawing tablets and stylii with virtual desktops, but have been prevented from doing so by latency.
USB redirection to cloud is an effort to overcome that problem. At VMworld 2015 last week Citrix demonstrated a WACOM tablet being used to drive Autodesk Mudbox, running inside a Citrix XenDesktop session. Even over conference-grade WiFi, and a trip over the public internet to Citrix's bit barns, the tablet and its stylus worked, complete with pressure-senstive nuances. So far, so niche. But Citrix thinks this could also in handy for applications like signature pads on credit card machines. Given criminals' liking for point of sale machines this sounds like a good idea, as a signature pad that sends data to a virtual machine in a nicely-secured data centre sounds rather more secure than signatures managed on a PC with a cash drawer attached. Citrix says “USB Redirection to Cloud will be part of a future XenDesktop and XenApp release.” At a guess, that might mean one of the “feature packs” Citrix drops at regular intervals or version 8 of the Xen products that, if Citrix sticks to its usual release practices, will land sometime in 2016.
Continuous lifecycle London 2017 event. DevOps, continuous delivery and containerisation. It is cast as the silver bullet to overcome today’s challenges in almost every area of our technical life: The Internet of Things. IoT is being positioned as a one size fits all technology, where sensors are connected to the cloud to give us more comfort, pace, insights and – most thrilling – an unprecedented data pool to predict what is to come. However, Smart Homes and Smart Rigs are a little bit different. An incorrectly adjusted thermostat resulting in a 0.5 degree warmer home temperature, may cost you 50 USD more per year. In contrast, what happens if you fail to predict that a blow-out preventer will malfunction? Well a little bit more, something around 65, but with nine zeros to the right. Following the same analogy, what do you measure with a smart thermostat? Average home temperature, either in Fahrenheit (centigrade in my world and maybe kelvin if one day I live in outer space) and that’s it.
What do you measure on an oil rig: thousands of things? And what you most importantly need to understand is the context of those measured things! It is critical to know where the measurement is coming from and how it is related with the rest of the rig’s measurements. What sensors are on the same type of mud pump and what is the feed to those pumps? And of course how do you know that the same type of mud pump is in operation on multiple rigs and how do you compare across your hundreds of oil wells around the world? So, to understand comfort and safety with respect to IoT in Oil & Gas, let’s first talk a little bit about the “internet” and then about “things”, citing some key examples. Get Connected to the Vastness While surfing on our mobile phones at blazing fast LTE speeds, we tend to believe that the whole world and everything in it is covered. But very often, the remote areas of interest for Oil and Gas exploration and production need more manual solutions.
Seismic surveying is one of the oldest and most widely used methods of exploring hydrocarbon reserves miles under the earth’s surface and ocean floor. Recent innovation has produced new IoT-enabled wireless seismic sensors, that could support this industry to not only find new reserves, but also recover more hydrocarbons from existing ones. This new technology has been tested in remote areas of the world from the North Slope of the Arctic to the mountains of South America. Obviously we don’t connect these sensors directly to the Internet and pump data into Azure IoT Hub. First and foremost, you don’t expect LTE coverage in the North Slope of the Arctic (which might be why we don’t see lots of polar bear selfies on Instagram). One of the techniques in place is to use a vehicle with WiFi antennas to read the accumulated sensor data while moving across the remote terrain and subsequently bulk upload the gathered data to the processing center. Another option could be a mobile datacenter mounted on a truck to provide a mobile hotspot to collect the data directly from the sensors.
When all other options are infeasible, the data is transferred the old-fashioned way via sneakernet: first stored on hard disks and helicoptered to the data center. For instance, Microsoft Azure has a service where you can send an HDD to be loaded to your blob store. Overcome High Latency and Low Speed Another challenge in the Oil & Gas world is to manage the life of workers stationed on offshore oil and drilling rigs and floating production vessels. This needs specialized fleet service organizations not only to provide accommodation and food, but also to ensure safe and efficient operations aboard the vessels themselves. In the past, fleet management companies had to collect time-sensitive data by sharing it via telephone conference or sending staff to the remote installations. For routine data transmission, oceangoing ships and platforms anchored to the seabed, must rely on satellite networks, which are often slow and prone to outages. Kongsberg Maritime, a provider of dynamic positioning and automation systems, looked for better solutions to help onshore teams manage equipment performance and monitor the safety of offshore installations.
In collaboration with Attunity and Microsoft, Kongsberg created a data-transfer solution that efficiently relays only the changes in the overall set of captured data. If the satellite connection is interrupted during transmission, their system automatically resumes data transfer once it is restored. Solve the Hidden Depths of the Things in Oil & Gas The second most important piece of an IoT solution is obviously the things. In the Oil & Gas industry, this means tackling some very challenging tasks: Handle mission critical and complex measurements: as we said in the beginning, while a thing can be a thermostat in your home with a single temperature value, industrial things in the Oil & Gas industry include electrical submersible pumps, fracking trucks, and oil wells with hundreds of surface and sub-surface measurements (tags) that are sometimes updated multiple times per second. The other noteworthy aspect of these things is their mission critical status, where malfunction and subsequent downtime can cost Millions or even Billions of USD in lost production, not to mention the potential danger to the safety of personnel and the environment.
Gather context to understand measurements: As important as the measurement of our things itself, are the location, type, connections and dependencies with other things and their measurements. For instance, we should know that the mud pump of interest is running on well XYZ and its respective inlet and outlet pressures, temperatures and flows should follow certain patterns. We also need to follow common naming standards, units of measure, etc. to be able to compare and correlate with similar mud pumps, in order to avoid silo/localized optimization. Information models that incorporate industry standards and hierarchies are decisive – whereas, a measurement by itself has little value without context. Standardize to manage data protocols: Standardization of communication protocols was mostly an afterthought for the manufacturing industry’s equipment and software systems. Hence there are around 450+ protocols (as of my last count) with which one needs to communicate in order to collect the data needed from these things.
While the OPC Foundation has gained traction in this space, as seen in the thing-manufacturers who have adopted this protocol, there are still many more to cover. Software solution providers, such as OSISoft, have built hundreds of adapters to extract data from these different protocols and store it in their PI historian. With PI to PI Cloud Connectivity, these data measurements can then be easily ingested to Azure for further processing, analysis and visualization. Enhance security with gateways: Historically, Operating Technology (OT) networks were proprietary and isolated from the Internet, and therefore, enjoyed the benefit of being secure by default. Recently, we’ve witnessed incidents such as the Stuxnet attack where great damages can occur if OT and IT networks are connected. This is probably the single most critical factor affecting the widespread adoption of cloud based solutions. Instead of directly connecting the things to IT networks, gateways with data diodes or firewalls are being deployed as secure channels.
Equally as important is addressing the security capabilities of the classical OT protocols. Some of the newer IoT-born connectivity protocols, built for smart homes and similar consumer devices, will need to be carefully adopted in the industrial realm. Take the last mile with existing infrastructure: Edge Processing has gained more importance for taking local actions, when sending all of the data to the cloud for processing and analysis is infeasible due to security, network bandwidth limitations and autonomous operation requirements. There are a lot of these use cases with brownfield scenarios where infrastructure is already in place to collect and process data for SCADA, DCS, PLC, RTU, etc. This infrastructure can provide the edge processing capabilities such as protocol conversion, data aggregation, store and forward, encryption, compression, and more. Azure can then deliver incremental value with new capabilities such as hot path risk pattern recognition and notification as well as long term storage and predictive analytics capabilities using machine learning.
Benefit from Azure as a Lego set: I often describe Azure as a Lego set. I still like to visit the Lego store, especially to explore the new NXT versions. The magic of Lego is that with standard pieces you can build anything you want. You are only bound by your imagination and creativity. The only custom parts you need, for instance, would be in the case where you want to build the Battlestar Galactica. The Azure platform is virtually the same, in that there are hundreds of standard Lego blocks. In addition, the Azure Marketplace provides custom blocks for various industries. For instance, if you want to build an IoT solution and need a cloud scale historian, OSIsoft has released their product running on Service Fabric. If you want a reservoir simulation, you can run Schlumberger Intersect in Azure. With the ever growing number of custom blocks, one day soon, you can build your own Battlestar Galactica. Fasten your seat-belt and get ready for lift-off. In the next posting, we’ll discuss the five key steps to achieve the Internet of Your Oil and Gas Things.