RAIL INDUSTRY FOCUS
UPGRADING A RAILWAY NETWORK
An upgrade to Belgium’s entire railway network to state-of-the-art technology needed to be carried out at minimal cost and at maximum speed. Stratus Technologies explains how the requirements were met
K
eeping commuters safe and on time is crucial for railways all over the world,
and a signal, level crossing or track change failing could cost lives. But while railway infrastructure needs to be maintained and sometimes upgraded, this needs to be carried out with as little downtime as possible. So when Infrabel (the manager of the
Belgian railway infrastructure), Siemens (who provide the application and platform support to Infrabel) and Stratus Technologies project managed an upgrade to Belgium’s entire railway network to state-of-the-art technology, it was carried out within just one hour of planned downtime per location.
THE OBJECTIVES The railway company’s objectives were to upgrade the infrastructure to meet future safety standards without compromising availability at any point of the process. The suppliers – Infrabel and Siemens – were required to find a way to implement such an upgrade at minimal cost and maximum speed. Erik Devriendt, senior consultant with
Siemens industry solutions division, takes up the story: “Like most developed countries, Belgium’s railway network is controlled from computer-based control centres. The safety of the entire country’s train traffic depends on the electronic interlockings in these control centres being fit for purpose. “Siemens works with Belgium Rail’s national
infrastructure management organisation, Infrabel, to ensure this. We provide a purpose- built operations and control system called EBP, which connects to these interlockings, presenting a human interface to the operators at the control centre. The EBP presents graphical layouts
updated in real time with the status of all signals, points and tracks, as well as the position of each train. The operator can create and change planned train trajectories via the EBP, which updates the corresponding signals automatically. If any element of the infrastructure malfunctions, the system guides the operator through the actions necessary to keep the system safe, no matter what. “There are 27 of these EBPs located around
the country to date, each hosted on a Stratus Continuum server. With newer, faster, easier to maintain technology now available, the time had come to upgrade and increase the total number of systems to 31. Doing so with zero impact on the running of the railway network at minimal cost and maximum speed was the brief.”
UPGRADING WITHOUT IMPACT “Upgrading an entire country’s railway infrastructure, causing zero impact to its customers, is a tall order,” added Stratus project manager, Brent Thomas. “Siemens, together with Infrabel, achieved this thanks to their careful choice of technology and meticulous planning, testing and project management.” Devriendt commented: “Although we reviewed
other new fault tolerant infrastructures, such as HP NonStop and Stratus Linux-based ftServer systems, Infrabel chose to stick with Stratus and to go with the top range V Series.” They knew Stratus service level agreement and 100% uptime assurance was reliable. Going with V Series meant easy porting of the existing VOS EBP application and provided additional reliability. Replacing 27 and adding five new systems
scattered throughout a country takes time, and project managing this task meant taking into account:
/ DESIGNSOLUTIONS
• The need for extra power and cables • The need to perform the updates during ninety minute slots available only two nights a week
• The need to use already overburdened system administrators’ resources “The extra system performance of the V Series
compared to the Continuum system was very useful here. It will also be used for additional EBP functionality and supporting larger EBP zones in the future,” added Devriendt.
MANAGING SWITCHOVERS Devriendt explained how the switchovers are managed: “We install and boot the new V Series with the Continuum system still in operation. We install the necessary software and read- only configuration data on the machine before switchover. During preparations the V Series machine has a different IP address, so that it is reachable via the network but does not interfere with the operational system. On the switchover moment we stop the application on the Continuum system and copy log-files and critical data over to the V Series. “New log-files are created each month,
so when the switchover happens at the beginning of the month we limit the amount of data to be copied to less than 100 Mbytes. We then disconnect the Continuum system and shut it down, changing the V Series IP address to the former address of the Continuum system and rebooting the V Series automatically and performing the necessary tests. This migration pattern works very smoothly and we now have it down to under 60 minutes per server.”
Stratus Technologies
www.stratus.com
DESIGN SOLUTIONS | OCTOBER 2020 33
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44
