For context, this post is being written on a beautiful sunny Wednesday afternoon here in Carlisle. It also happens to be near the beginning of the UK lock-down due to the Coronavirus pandemic which struck in March 2020. Businesses all around are being as helpful as they can be, and it’s amazing to see that spirit come out from the words business leaders. From startups to multinationals, we’re here to help. So today, we’re talking about Managing Your Cloud Resources In Times Of Crisis
Why You Should Care
Lots of folks think of the cloud as a nebulous entity, with no physical presence, and no limits. That’s a fair thought process to have when all most people see is the shiny marketing output of cloud providers. But the reality is, there are datacentres all over the world serving the cloud capacity we all rely on every day. Find out more in our previous post, How The Cloud Works.
Businesses and individuals move workloads into the cloud literally every day. The cloud needs to expand to keep up. What that means in reality is that cloud providers need to put more and more servers and other capacity into their datacentres. But if they add too much capacity, they have under-utilised assets, and don’t make enough money overall. It’s a tight balancing act.
But what happens when cloud providers can’t get hold of new servers? What if the places that servers are usually manufactured are no longer manufacturing? What happens when the datacentres themselves are running on skeleton staff (onsite staff at least).
Capacity Shortages
Yesterday, ZDnet reported that Microsoft Azure was throwing up errors to some users in the UK South region. The errors regarded a lack of capacity, restricting their ability to create new cloud resources. Put simply, there just weren’t enough servers to go around.
It’s not the end of the world though, there was still capacity in other areas you can use. The nature of the cloud is such that it really doesn’t matter where your servers happen to be running. However, it is still a bummer to be brought back to reality, and for your cloud provider to have to say no. But let’s not panic.
The Toilet-Roll Fiasco
First thing’s first – here in the UK we call it toilet roll. Not toilet paper. Or if you happen to be Mary Poppins, you probably call it Loo-Roll.
Anyway, my pedantic nature aside, we all know what happened when people went daft buying up more toilet roll than they could possibly use in a year. There was none left for folks who had actually ran out of it. So let’s not do that with our cloud resources, yeah?
That said though, if you have capacity which you frequently turn off and on again during peak times, it might make sense to leave those resources running at the moment. Use this as an opportunity to focus your IT strategy. – Do you really need those servers if they’re switched off half the time?
If you do, fine, but make sure you’re paying for your resource as efficiently as you can be. We have a blog post on the way, all about reserving cloud capacity, so stick around for that in the coming weeks. In short, commit to running certain services for a year or longer, and receive hefty discounts. This helps the cloud providers plan future capacity needs.
But also, maybe clear out some stuff you don’t need and make some room for others. Since everything in the cloud being pay as you go anyway, it makes perfect sense to have a good strategy for removing old and unused cloud resources. A good place to start is to check the tags assigned to your assets. And if you’re not using tags – start using tags! Oh how we love a good tag.
Meet You In The Winchester
This will all blow over soon enough. We all have the amazing people in our health services, police and government to thank for their determination and continued efforts to get our world back on track. If all we have to worry about is having to move some of our servers to other cloud regions, we’re alright aren’t we.
There are plenty of resources here in our Blog archive to help you understand how all this works, so stick around and have a little read.
