Computer-aided design (CAD) and computer-aided engineering (CAE) often use extremely large files, which make remote working and collaboration difficult to achieve. Most businesses are able to work remotely with laptops, smartphones and tablet PCs, but the processing power required to create, store and share CAD files over a wide area network is often prohibitive. Yet with the power of the internet they can work remotely and collaboratively.
The key challenge, because the files they send to their colleagues and CAD partners across the globe are so large, is about how their IT teams mitigate the effects of latency in wide area networks (WAN) to enable them to work uninterrupted by slow network connections. Strained WAN resources aren’t the only issue that concerns them. They need to deploy remote access control technology to protect their data as it flows across the internet, and from cloud to cloud, to ensure that only authorised individuals can work on any given CAD or CAE projects.
In essence, the internet and more recently cloud computing has become an enabler of remotely situated design, manufacturing, construction and engineering teams. Not only can they share their skills, knowledge and expertise, but also their data. Cloud can handle the peaks in storage and computing demand, but organisations must be able to get it up and down quickly in order to benefit from the potential cost efficiencies and the infrastructure agility that it can offer.
“With the ubiquitous access to the internet it is now possible to gather data from every part of the world and bring it back to a central hub for analysis, and you can design an aircraft or a car in one country while manufacturing it in another”, says David Trossell, CEO and CTO of data acceleration company, Bridgeworks. He points out that this means a huge amount of data is constantly being moved around and that all of the data is logged.
It’s about time
The Engineer says in its June article, ‘It’s About Time – Evolving Network Standards for the Industrial IoT’, “The Industrial Internet of Things (IIoT) promises a world of smarter, hyper-connected devices and infrastructure where electrical grids, manufacturing machines, and transportation systems are outfitted with embedded sensing, processing, control and analysis capabilities.”
The article recognises that latency can still be a problem, and it claims that:
“Much of today’s network infrastructure is not equipped to handle such time-sensitive data. Many industrial systems and networks were designed according to the Purdue model for control hierarchy in which multiple, rigid bus layers are created and optimised to meet the requirements for specific tasks. Each layer has varying levels of latency, bandwidth and quality of service, making interoperability challenging, and the timely transfer of critical data virtually impossible. In addition, today’s proprietary Ethernet derivatives have limited bandwidth and require modified hardware.”
The article adds: “Once networked together, they’ll create a smart system of systems that shares data between devices, across the enterprise and in the cloud. These systems will generate incredible amounts of data, such as the condition monitoring solution for the Victoria Line of the London Underground rail system, which yields 32 terabytes of data every day. This Big Analog Data will be analysed and processed to drive informed business decisions that will ultimately improve safety, uptime and operational efficiency.”
Commenting on the article, and specifically about the London Underground example, Trossell says: “Everything is real-time, and so the question has to be: How can we get the data back as fast as possible to analyse it and to inform the appropriate people. Some elements of this task may be in-house first for a quick exception analysis?” Some of this data may then be pushed to the cloud for further in-depth analysis by comparing present data with historical data to see whether anything can be learnt or improved from a maintenance and service perspective. With an unimpeded network, big data analysis from a wide range of data sources is possible, adding the ability to gain insights that were once not so easy to obtain.
From IoT to IIoT
Trossell thinks that the broader expression of the Internet of Things (IoT) is just one of the current buzzwords that everyone for a variety of reasons is getting excited about. “Most people think of this as their connected fridge, the smart meter for their utilities or the ability to control their heating system at home, but with the ever increasing diversity and the decreasing cost of sensors for industrial use, the term takes on a new level of sophistication and volume when applied to industry”, he explains.
In industry, IoT gives birth to IIoT, which involves monitoring the performance of complex machinery such as gas turbines, aircraft, ships, electrical grids and oil rigs. So it’s not just about a diversely spread group of CAD engineers working collaboratively across the globe. “IIoT has never been so diverse and in depth with vast amounts of data being created every second. To put this in perspective, each Airbus A350 test fight can received measurements from 60,000 separate sensors”, he claims. That’s a phenomenal amount of data that needs to be transmitted, backed-up, stored, and at some point it needs to be analysed in real-time in order to have any value.
“An example of this is a company that has developed a system where the aircraft technician can download the data from the black box flight recorder, send it over the internet where it is analysed with artificial intelligence for anomalies exceptions and then passed to an expert for investigation”, he explains. The benefit is that this approach can engage several experts, for example from an air transport safety board and involve manufacturers, across the globe to find out, unusual pilot activity, or sensor data can be collated to enable an airline to reduce maintenance and unplanned outages and possible safety implications to a minimum therefore improving availability and profitability.
“However, just like the consumer internet, moving vast amount of data across the internet has it challenges – especially when the data may be half the world away”, he warns. These challenges include increased network latency due to the teams working at a distance over a WAN, and potential security breaches. He adds: “Moving files around between various data silos can be inhibitive even over a LAN – the cost of 10Gb networks are dropping considerably, but with WANs the problem is about moving data over distance because of latency.”
Yet in spite of the gremlins posed by security threats and network latency, there are many companies around the world that are established virtually thanks to the internet. They are often specialists in their chosen disciplines, and each of them can add a bit to the whole picture, but Trossell believes it’s no good to collect or generate data if you can’t use it to encourage and enable the collaboration of globally dispersed multi-disciplinary teams, to allow for innovation and for the creation of efficiencies. The data – including sensor data – must get to the right people at the right time if it is to add any value, but latency can prevent this from happening and latency can turn invaluable data into redundant and out of date data that adds nothing of worth or merit.
Companies investing in IIoT and remote working therefore need to protect their businesses by investing in solutions that can mitigate the impact of network latency while enabling data to be securely sent at velocity between the various data users and analysers. With smart systems, the challenges can be harder to overcome because in a traditional Purdue system data flows up and down the model, but Trossell says that smart systems and IIoT data tends to flow in all directions like a web. Being smart is also about mitigating latency and reducing the potential threats to data. With this in mind, Trossell offers five top tips that could ensure that your company gains the most from its data:
- Remember that the two biggest killers of performance in WAN is packet loss and latency. When you have them together then you will suffer massive performance hits.
- Adding bandwidth to your WAN will not necessarily increase performance, but it will increase costs!
- Marshal and consolidate data if possible rather than allowing lots of individual streams as this is a more effective use of WAN bandwidth.
- Use a product such as PORTrockIT to accelerate data transmissions, and use applications that pre-compress and encrypt data before it is sent to the WAN.
In essence, by mitigating latency and improving data security, industrial organisations can maximise the potential of the industrial internet. With the growing amount of sensor data, and the growing need to work collaboratively with remote teams across the world, these challenges are going to become more and more prevalent and obstinate. Industrial organisations therefore need to act today to ensure they protect their businesses well into the future, to enable them to participate in the industrial internet of things, and to allow them to benefit from real-time big data analysis right now. In other words, there is no point in using smart technology if you aren’t being smart with it too.