There was a time when a single kilobyte was revolutionary. Then the megabyte, then terabyte, and so on. Even a decade ago, a company holding a petabyte (1 thousand terabytes) of data was a lot; it was likely that only very large or data-rich businesses would have so much.
Nowadays, a petabyte (one thousand petabytes) is nothing to a business. Some are storing data far beyond that. The tricky part isn’t producing all that data. It’s knowing where and how to store it. Data provides so much value nowadays, but it’s useless if it’s not being stored properly.
It’s very hard to say just how much data the world is producing every day. It’s been quoted that Facebook alone produces 4 petabytes of data every single day, most of which is people uploading their pictures and videos. Or consider something like YouTube, which allows people to upload an infinite amount of HD video. All of this has to be stored somewhere.
The amount of data being created is exponentially rising due to the Internet of Things. Loads of devices are connected now, including watches, fridges and washing machines. A single connected car produces 120 terabytes of data a month. Wearable devices produce 335 petabytes of data per month in 2020. These are huge numbers.
The IDC predict that we will have generated 175 zettabytes by 2025. They also say that 59% of the shipped storage capacity will come from hard disk drives.
When you consider how long hard disk drives have been around, and the fact that some businesses still store on tape, it’s unlikely that the storage landscape is going to look drastically different in five years’ time.
Market research firm TrendForce estimate that, of the hard disk drive market, Seagate own 41%, Western Digital 38% and Toshiba 21%. While hard drives are dropping from the overall storage market share, the capacity of the drives is on the increase.
In the meantime, Seagate and Western Digital are soon looking to ship 50TB and 100TB drives. Western Digital plan to use energy-assisted perpendicular magnetic recording technologies in order to boost the areal density. They are also working on an architectural change called Zoned Storage, which standardises the randomisation layer and creates distinct namespaces – the overall result being a reduction in DRAM resources for each NAND drive.
But new technologies are being developed and these will begin to enter mass market as they become easier and cheaper to produce. Companies are looking into how to store data on things like glass and DNA.
The current rollout of 5G is also contributing to how data storage can adjust in relation to access patterns. “In addition to the core applications like video, we have connected endpoints happening,” says Yusaf Jamal, Western Digital’s senior vice president of devices and platforms. “As technology is adapting to how we live in our society, how our health is transitioning to digital health, how are factories are transitioning to smart factories, we have these connected endpoints generating more and more amounts of data that we have to consume.”
How Can the World Cope With Exabytes of Data?
No comments yet. Sign in to add the first!