You might have seen my recent post about a new managed services client for Ideal. A year in the planning, the first phase of the project spanned 400 devices in 31 sites, scattered across 18 countries. It was a big job, and there's more to come, so it was essential that we could deliver the excellence we expect from ourselves. Most importantly, it was imperative that there was as little disruption as possible to our client's network, and hence its business.
But how do you configure and onboard that many devices, spread across so many timezones, from a managed services centre in England? Further, with most locations having no on-site networking expert, how do you ensure that there are no mistakes?
I was confident that our in-house experts would deliver, and indeed the first phase has been a resounding success. How was it done? Over to network support engineer Ryan Clawson...
One of the first stages was for our engineering team to conduct a full audit, giving us a comprehensive list of what we're bringing under management. We're doing this for all sites by comparing customer-supplied inventories against our remote audits, after which we can build a proper map of what's present at each location and how it's currently configured.
A core part of the project was to build a centralised authorisation server to control access and provide an audit trail for logins and changes. Finally, another key stage in our preparation was the creation of a global standard configuration for network devices - a job for solutions architect Richard Harvey. I took this and essentially dropped it into a scripting tool called Solar Winds.
For each site, the first change I make is to move each piece of network hardware over to the new authorisation server. Next, I can log on and run my script. The key thing here is that the tool queries the device to discover variables like the exact model and revision, then asks me a few simple questions such as the device location, who the primary technical contact is and so on. Then it compiles the variables into a script and fires it at the device, automatically configuring it to our global standard.
It's an incredibly efficient approach, using which we can apply the necessary changes to a large batch of equipment at a time when it won't cause much disruption to the business: at one stage a colleague and I were deploying to 26 devices at once. Another key benefit is that we're automatically rolling out a very carefully prepared and tested configuration: this way there's much less likelihood of error. Afterwards I check that every device is onboarded correctly - we haven't had a single failure.
Need help with your business challenges? Call us on 01273 957500, or get in touch online.