A copy in the cloud: why and how?
We’ve been storing our partners’ audiovisual content in separate physical locations ever since we were founded. But what difference can cloud storage make in terms of cost-efficiency? We found out: specifically for backing up the low-resolution copies, it leads to reduced direct storage costs and lower operational overheads.
In order to guarantee our partners’ audiovisual content is kept safe and secure, we store various copies in different locations. Saving high-resolution copies means we ensure the very best quality is preserved in a sustainable way. This is a perfect and necessary solution for indefinite storage and preservation, but not suitable for fast access to files over the internet. That’s why we also save low-resolution copies – smaller derivative files that can be used to stream content to our partners’ platforms or to our own platforms, such as The Archive for Education and Catalogus Pro.
Two copies of these low-resolution files are saved in two different locations. Specifically, until now, a copy has been saved in each of our data centres in Ghent and Oostkamp. One copy is active and can be retrieved immediately, and the other isn’t active and serves purely as a back-up in case the active copy becomes lost or damaged – in a fire, for example. It’s very important that the back-up copy can be activated quickly in the event of any problems with the first, active copy.
What's new?
Technology is always on the move, so we keep our finger on the pulse when it comes to new storage solutions. Cloud storage is one of these new solutions – it means you’re not storing data on servers that you pay for and keep operational yourself, but rather buying storage as a service provided by a third party. So you’re no longer responsible for investing in your own servers and hosting, or their operational monitoring. Another significant advantage offered by this solution is flexibility – with cloud storage you only pay for what you use, so you don't need to buy hardware or software in advance of any predicted growth. The storage capacity (and with it the cost) increases in line with the archive’s expansion.
In order to see if this solution would add both functional and economic value for us, we compared the price of storage services in the cloud with the cost of our own storage system – and ultimately came to the conclusion that cloud solutions are more expensive than storage in our own data centres, except in one specific case: backing up the low-resolution copies.
Manipulations of this content are limited, which is a good thing considering every use or manipulation of a file comes at a cost.
This means we could make a saving by migrating files that are actively used to a more affordable type of archive storage service. These files are then no longer immediately available, but can easily be re-activated as required.
In this scenario, the second copy acts as a back-up and migrates to the cheaper storage solution after 30 days. Further analysis showed us what the impact would be if meemoo needed to use this back-up: can all files be replaced fast enough? And would the cost for this be acceptable? Even when we consider the worst case scenario, whereby we need to activate all the back-up files – which is highly unlikely over a four-year period – this type of cloud storage results in savings that are sufficient to justify it.
We then drew a couple more conclusions about usage: activating the copy ultimately leads to a significant added cost, but it remains within the limits of what’s acceptable. Furthermore, meemoo can decide not to migrate certain collections to the second type of storage if they need to be back online quickly, so they’re immediately available at all times. With all this in mind, we decided to start working with the new cloud solution.