The Data Storage Forecasting Conundrum
From my time at a data-hungry media agency, one of my biggest challenges was storage forecasting, which always preempted my capacity purchasing decisions.
CFOs need pinpoint accuracy, as IT expenditures must be projected and planned for while companies strive to balance profitability in support of optimal financial health.
Just as important to my department was making sure we never ran out of storage space or risked impacting productivity (which would violate an IT prime directive!).
Enter the variables…
But storage growth isn’t always predictable, especially for media agencies.
Sure, I could use some trend analysis to track growth over time and continue charting a line that future milestones should hit, but that rarely proved accurate.
Applications from companies like Adobe keep adding file features that can greatly increase average file sizes overnight! .psb anyone?
In addition, the media-richness of customer deliverables has grown exponentially. The outputs are so cool, but they can eat storage blocks like Cookie Monster eats… cookies.
And don’t get me started on, “Our new customer has provided us with a complete archive of work from another agency, and we need to store it.” A sudden 100TB data packet is great at throwing a wrench in your forecasting.
Right-sizing your storage solutions to meet current and future needs without looking like a money sieve is a special art. And in my case, it resembled more of a Jackson Pollock than a Rembrandt.
What Caching Solves
Like a gift from tech-heaven (which I like to assume is something Gene Roddenberry created), I came across the global file system and the magnificent concept of caching.
I remember thinking: If only 6% of my entire dataset has actually been accessed in the last 90 days, why the heck am I storing it all the same way?
Traditional NAS wasn’t the trusty old certainty I thought it was; it had become the biggest piece of my architecture overripe for serious disruption. An anchor around my feet, tying my organization to a hamster wheel of hardware refreshes, futile capacity planning, and a black hole of cognitive overhead that I could be spending somewhere far more exciting.
The Exponential Benefits of a Global File System
The next generation…
So, let me get this straight. With a global file system, I can:
-
Present the entire file system to my users, exactly as they’re used to, without any impediments to file retrieval?
-
Retain all of my file and folder security, and provide a cutover that’s completely imperceptible by my users?
-
Ensure that all accessed data (hot) is high-performance on my local network while bringing back files from object storage cache is fast and automatic?
And since my local storage needs are now a fraction of what they used to be, I’m going to save a ton of money, never be faced with running out of space, and gain ultimate flexibility over who provides my object storage and where it resides.
Talk about meeting cost, reliability, and data sovereignty requirements!
Of course, I played it cool in front of the sales team and pretended this was all rather pedestrian while my mind began calculating the actual scope of benefits I could unlock 😉
Goodbye Storage Guesswork
At that time, I thought NAS was just the way. Turns out it was only my way to spend more money on storage.
I get it, “the devil you know,” and such. But caching is transformative for long-term spending. I dare anyone to show me a data access report that doesn’t indicate that the vast majority of their data belongs in a far more cost-efficient place, without changing how people access it.
For this guy, NAS will not live long or prosper. Perhaps it’s time for you to “make it so” as well?
Thanks for reading! 🖖
