Save Money – by buying more disk

How much does it cost to manage your data?

In an early job, I was making a projection to purchase more disk to expand capacity on our Exchange mailbox servers. At the time, disk was relatively expensive, and the capacity that I had projected for 3 years would have also required a new DAS shelf and RAID controller – so the request for funding was knocked back. However, I was armed with more than just the techie’s desire to buy more kit, and returned with an analysis;

For each mailbox that was close to being full (there was a ‘generous’ 500 MB limit per mailbox, that nearly 50% of the organisation were regularly hitting), I calculated the wasted time it would take for people to spend on cleaning up their mailbox – and the lost productivity that this incurred (lost productivity needs to be tripled – the work that could have been during the hour doing a non-core task needs to be performed later, which also impacts on productivity for a further hour). I also calculated the impact of users simply dumping their mailbox into a PST file that would be stored on a fileserver (slightly cheaper disk than Exchange server disks, but still a cost) – or the risk that the PST would be stored on a local computer. I also estimated the wasted time in looking through PST files to find archived emails, and the risk that these PSTs would become corrupt or lost – or the impact of erroneously deleting emails that would be required later. Although most of my calculations were based on estimates and averages, I pointed out the key finding;

Not buying more disk would cost more.

The productivity of people would be impacted, the risk of losing information (or simply making it harder to find) would have more of a business impact and cost than buying some hardware and tapes.

The cost of storage in the Cloud era

Sticking with Microsoft Exchange for a while – the capacity of mailboxes for Exchange Online is 100GB per user in 2018, it was 50GB in 2016, and in 2013 the maximum capacity was 25GB – so Microsoft is constantly adding more storage capacity for users. Ignore for the moment that this is justifying my above calculations, and instead focus on how the limits are constantly moving – where there were technical limits in the past on having mailboxes larger than 2GB, the technology is changing the goalposts.

Other platforms are providing “unlimited” storage – and at a very low cost. The Microsoft OneDrive for business platform still needs a little manual intervention around the 25TB mark, see here.

But other technologies also offer “unlimited” storage, such as Amazon S3 object storage, where the consumption based costs are also very low https://www.amazonaws.cn/en/s3/faqs/ – all an indication that the industry is racing to zero.

Searching and finding data then becomes the challenge – people still have a habit of browsing for data, where they look in filing hierarchies. However, when you store everything – you need to have an easy way to find what you want. You will know what I mean if you have you ever tried to find a particular photo, from a folder of thousands of sequentially named files. The other extreme can also be frustrating – I’ve worked in PRINCE2 influenced project teams, where they create intricate folder hierarchies for storing documents – and most of the folders are empty and a lot of navigation is required to find the right file.

The drawbacks of easy access storage

Many years ago, IT administrators would fight against the users who would upload their entire computer contents to file servers “just in case”. We would frequently find folders called “C_Drive” and multiple copies of full Windows installations filling up space on the file servers. Are we going to have this with unlimited storage areas in the Cloud? Probably – but then it will be someone else’s problem…

What about on-premises data storage? What if it isn’t someone else’s problem? If you don’t buy more storage capacity and require that people take time out of their productivity to manage their own data, you run the risk of driving people to Shadow IT. It’s cheap (or free) and easy for anyone to get a few gigabytes of storage (click here to get a free 5GB of OneDrive) – which is then available to them from any device, and allows sharing with colleagues. Even slightly savvy users are, or will be, creating business files on private storage – and other users will not bat an eyelid over it, and probably join in and use their own storage services.

If official business capacity is limited, Shadow IT starts – users know they can get what they want directly

https://www.hpe.com/us/en/insights/articles/interested-in-consumption-based-it-here-are-5-questions-you-must-ask-1808.html. Which battle are you trying to win? Saving costs on hardware, or improving business resilience and collaboration?

Summary

In the past, IT administrators needed to control data sprawl due to the costs of disks and backup, and technical limitations. Practices of creating filing hierarchies and browsing for data are in-grained and replacement practices are not mature. With the technical barriers and storage costs changing, and the availability of free capacity easily accessible through Shadow IT, IT departments will need to service this need – or acknowledge the risk of losing company data on unofficial systems.