Server virtualization for manufacturing and industrial environments might be a hot topic lately, but many of our clients have been reluctant to jump on the bandwagon just yet. While everyone loves to espouse the benefits – better application life-cycle management, less space requirements, readily available backups – at the end of the day it’s still scary to move all our applications to a virtual environment. To help give you a better idea of the trade-offs, not just the benefits, we have put together a list of the top pro’s and con’s of server virtualization for industrial plants:
- Reduced Space Requirements – by allowing you to consolidate applications onto one or two servers, less space for computers, servers and IT hardware is required. Space in a manufacturing plants is already highly competitive, so freeing up more of it for other plant needs is a plus.
- Lower Hardware and Utilities Costs – virtual servers require less physical hardware in your plant, less cooling requirements, and less electricity because most of the heavy application lifting is done in the cloud. Going from 8-10 or more distributed servers to 1 or 2 makes a larger difference in hardware and utility costs.
- Longer application life cycles – the less changes to your plant applications, server hardware, and operating system requirements, the less interruption and downtime for your operations. By visualizing applications, older operating systems can be run for longer, in a safer virtual environment, while server hardware is upgraded independently. This means you can actually run MES and other plant software for the duration of their life-cycles – 7-8 years on average, without paying for special support or risking failures from outdated operating systems not meshing with newer server hardware.
- Easier, safer expansion or upgrades with little or no downtime – you can create standard virtual machines that are pre-tested and simply copied onto new hardware when expansions or hardware changes are needed. This leads to more stable operations and changeovers. Virtual environments are simple copied, customized to new applications and switched over once pre-tested, with near zero downtime. You can easily expand capacity in the same manner, copying environments and then rolling them out in the cloud. Upgrades are seamlessly completed by copying the system state iteratively while the application continues to run. Shortly before a final copy of the virtual machine is ready for migration,only a brief application blackout (perhaps milliseconds) is necessary to sync the two environments.
- Backups and rollback capabilities – full image backups of configuration states, disk states, etc. are easily created and saved on more than one server in case of physical disaster. Additionally, many virtual server softwares offer rollback capabilities that allow you to rollback a server to a previous state in time where corrupted data or other issues did not exist in the case of an issue arising.
- Virtual server sprawl – because it is very easy to set-up and add new applications on a virtual server, it is also easy for applications to get out of control quickly. Without a dedicated management resource, multiple copies of applications, recovery points, or setups can start to proliferate in a disorganized mess.
- Speed implications – It is harder to optimize application availability and speed on shared virtual environments where programs are competing for less physical resources. Getting everyone to play nice and keep your systems running optimally will require help from a server virtualization application expert.
- IT Costs Can Rise – From tracking changes in virtual fees to requiring better IT support, the net cost of server virtualization might not be lower, especially if left un-managed.
- Virtual backups must be monitored – Backups become more important in this environment, especially when it’s harder to tell exactly what is being backuped where. While higher availability can be achieved by predicting hardware faulting and allowing migration of applications to backup servers or quick restarts, not all hardware failures can be predicted. You really need to understand what is being backed up and what isn’t, real restart times, and what worst-case rollover scenarios look like.
- I/O compatibility is required – before you jump into a virtual server technology, make sure your I/O are compatible with the new technology. This is the biggest culprit of instability and intermittent performance issues with new server virtualization, and may require upgrades in I/O hardware. SCADA systems are especially troublesome in this area.
- Security risks – Virtualization all your server data means extra security precautions must be taken for online connected environments. 1234 passwords aren’t gonna suffice anymore, and being lax about software updates, firewalls, and other protection methods puts more of your data at risk. IT vendors often provide update and protection services. but at an additional cost.
We find that server virtualization greatly benefits most of our customers that choose to implement it. People most often get into trouble when they don’t have a plan or have not consulted with an IT expert in server virtualization for manufacturing facilities. Because downtime is costly for plants, creating a thorough hardware upgrade, server migration, testing and rollover plan is a must for success. To learn more about our virtual server services for industrial plants, please give us a call at: 314-714-5055