Solution consultant extraordinaire Mark Andres has been working with product lifecycle management (PLM) for about 20 years both as a vendor and as a customer. He started 30+ years ago at Eastman Kodak in engineering and design, got recruited to EDA as their technical demo jock, and ended up in PLM in ’96.
In part three of our interview with Mark, he focuses specifically on the topic of how cloud providers manage PLM patches so companies can focus on their core business.
Arena: When it comes to managing enterprise systems, how do you roll out patches as well as test them?
Andres: In trickier implementations we can apply sometimes as many as 8 patches or more to make the system fully functional. The patches have to be installed in a very specific order and take many minutes per patch to install and then test.
Arena: What about the rollout of normal patches?
Andres: Again you don’t just roll these out. You install in a test environment and test and test and test. All these patches require ‘down time’, you can’t just take an enterprise system down at any time of the day during ‘business hours’, you have to schedule downtime that is ‘after hours’ and guess who gets to do that? Mr. IT guy, who probably had to work late the previous weekend doing the same for a different enterprise system.
Arena: Are there options to managing the rollout of these patches other than having IT guys work late at night instead of mobbing in the club VIP?
Andres: Of course there are other options, but that also comes with a cost. You could configure your system to have failover servers. In that instance, when you take one server down, everyone else is immediately redirected to the failover server. Patch that server, bring it back up and redirect everyone to that server, while you patch the other server. You see a trend here, more servers, more complexity, more cost etc.
Arena: While that may be serviceable for some SMB companies but not so simple for a global company, correct?
Andres: Yes. So as an IT department leader, someone has to come in late in the evening or the weekend, install the patch, go through a checklist of tests to be sure the system is available first thing in the morning.
It’s a funny thing to think of buying a product as a ‘best practice.’ But hey, cloud is in itself a ‘best practice’.
Arena: Talk to me about the benefits of the cloud when it comes to the topic of maintenance: backups, fail over/redundancy, security, etc.
Andres: Here’s what to consider when deploying a ‘big box’ PLM solution that is taken care of for you in a cloud-based PLM system.
When it comes to backups, customers need to ask themselves:
- What kind of backups do we need?
- Are there going to be issues with the operating system (OS) database and content?
- Are the scheduled backups going to be running reliably?
- Do you want hot or cold backups?
- How quickly and reliably can we restore our (system(s), database, content etc.) in the case of a disaster or outage?
- Where do these backups get stored and for how long?
- With regards to failover, how much down time can their business tolerate with minimal impact?
Arena: What are the questions companies need to ask with regards to security in their PLM systems?
Andres: Manufacturers looking for a PLM solution should ask questions like:
- Do we allow external access?
- Almost every company considers this since much of their manufacturing is outsourced. If so, how secure is this PLM solution?
- Can we prevent an outsider from seeing product data that we don’t want them to see?
For more information on the benefits of Arena’s cloud-based PLM solution, check out our key differentiators section.