Choosing the right storage strategy is becoming increasingly important as AI drives rapid data growth, making data classification, lifecycle management and resilience critical to controlling cost and risk. According to Mandi Pietersen, organisations should align storage technologies with workload requirements rather than relying on a single storage platform.
Storage isn't the problem anymore. Data is cheap to store and easy to find a home for, the real cost shows up later, in decisions nobody made on purpose. Video files balloon, business systems keep generating records, sensors and monitoring platforms never switch off, and now AI workloads are stacked on top of all of it.
The hard part isn't finding somewhere to put this data. It's knowing what the data actually is, how often anyone touches it, how sensitive it is, how long you're legally required to keep it, and how fast you'd need it back if something went wrong. Get that classification wrong, and no amount of cheap storage saves you from the bill, the compliance gap, or the slow recovery when it matters most.
"Dariel works with Azure, but also within SAN and NAS environments day to day. I no longer think of cloud and on-premises as rivals. In practice, a good storage design usually mixes both, and the workload decides the ratio," says Mandi Pietersen, Senior Cloud & Systems Administrator, Dariel.
Different users, different problems
Most individuals are storing photos, documents, videos, email and the odd financial record. An internal SSD handles day-to-day speed, an external drive gives you a cheap second copy, and a small NAS box ties a few household or home-office devices together.
Where people trip up is treating sync as backup. It isn't. If a file gets deleted, corrupted, or encrypted by ransomware, that change syncs everywhere else too, cloud copy included. Real protection needs three things: local storage, cloud storage, and a separate copy that isn't connected to either.
Businesses juggle a far bigger mix: customer and staff records, financial transactions, databases, contracts, logs, backups. "This mix is exactly why businesses end up needing file, block, and object storage side by side, rather than one tool for everything," says Pietersen.
Data centres scale the same problem up again. Virtual machines, databases, websites, AI training data, surveillance footage, backups, often for multiple customers, all at once. At that point, performance, redundancy, cooling, power draw and recovery time stop being separate line items and become one decision.
Where storage is heading
Storage has quietly become a lot more automated. Azure Blob Storage now offers hot, cool, cold and archive tiers, and lifecycle policies shuffle data between them without anyone lifting a finger - active data stays hot, older data drifts down to cheaper tiers on its own.
A few trends worth watching:
- NVMe and flash for anything latency-sensitive
- Object storage and data lakes absorbing unstructured and AI data
- Software-defined storage decoupling management from the hardware underneath it
- Edge storage handling IoT and video closer to where it's generated
- Immutable storage that simply won't let data be altered or deleted once it's written
AI is a big part of why storage demand keeps climbing and increasingly, part of how it's managed too. AI-assisted tools can classify data, forecast capacity, and flag unusual growth or activity that might signal ransomware. None of that replaces a person making the final call on deletion, retention, or anything touching confidential data. It just gives that person better signal to work with.
So which platform is right?
There's no single answer, it depends on what's being stored and what happens if it's lost. An individual mostly needs something affordable with a real backup behind it. A business needs file, block and object storage matched to how important each dataset actually is. A data centre needs SAN, NAS, flash, object and archive tiers all pulling in the same direction.
The businesses getting this right aren't the ones spending the most on storage. They're the ones who took the time to answer the boring questions first, what is this data, who touches it, and what does it cost us if it disappears. Ends.
About Dariel
Founded in 2001 on the principle of delivering solutions right, the first time, Dariel bridges the gap between human ingenuity and technology. Our strong client partnerships reflect a commitment to excellence and our consultative approach to software engineering makes us a trusted partner for innovative and sustainable tech solutions. Proudly independent, Dariel is part of the JSE-listed Capital Appreciation Group. https://www.dariel.co.za/
For more information:
Samantha Hogg-Brandjes | GinjaNinja | samantha@ginjaninja.co.za | +27-84-458-4857