Why SSDs are taking over in the data center
The reports of hard disk drives’ (HDDs) death have not been greatly exaggerated. That’s because solid state drives (SSDs) continue their takeover in consumer devices and the data center.
Sure, HDDs and SSDs each have their own advantages, but the latter is the more modern technology that continues advancing. As prices drop, SSDs are proving their advantages to end users and IT managers.
But there’s another huge motivation for large organizations to pick SSDs over HDDs: complex data. I’ll explain.
Not just traditional text files
Today’s 1s and 0s are a lot more valuable and complex than in the past. Organizations are storing the likes of videos, images, and audio, and other business-critical files.
One example is health care records. Hospitals, providers, and insurance companies aren’t just storing text files, but X-rays, CT scan images, and other information crucial to patient outcomes. SSDs offer an advantage in data centers, as they have no moving parts and therefore can access large workloads faster.
Not to mention their shrinking footprint. Traditional storage will always be constrained by its spinning platters and actuator arm; SSDs, however, are limited only by silicon. Engineers will continue to miniaturize these devices, and one day it may be possible to squeeze terabytes of storage to the size of a postage stamp. As the amount of critical data grows, the size of these drives continues to shrink.
Organizations, no matter the size or vertical, possess critical and complex data. As more Internet-of-Things devices come online, unstructured data will gobble up even more space. But organizations demand more than just speed – they need the assurance that their information is protected from power outages and equipment failures.
The true cost of SSDs
No doubt, HDDs and SSDs can both malfunction. But their failure rates paint a clear picture about which is more reliable.
Because of their moving parts, HDDs possess an annual failure rate of about 5 percent; SSDs, however, tend to perform more effectively for longer, with an annual failure rate of less than 0.44 percent. That’s in part because they lack the platters and read/write arm that easily malfunction, especially in end-user devices like laptops, which tend to receive more bumps and bruises over their lifetime. In fact, we’ve found the replacement ratio for HDDs to SSDs to be around 20 to 1.
SSDs’ low failure rate also affects the bottom line. Their cost curve is trending downward, approaching cost equity with HDDs. Plus, their reliability can mean IT has to buy fewer replacements than with traditional HDDs.
In addition, SSDs’ lack of moving parts means they draw less power compared to traditional storage, extending the drive’s life and reducing the total cost of ownership. Some SSD technologies will sense when a power shortage is imminent and save data, and then cycle into a lower power mode.
From the data center to driverless cars
SSDs can be applied in other applications, too. Think of a driverless car – it’s processing and storing data while driving over potholes, speed bumps, and uneven asphalt. From handling complex data to the need for more stable and reliable storage, SSDs fit the bill.
That adds to their advantages in data centers and consumer devices, where they fail less often than traditional HDDs and draw less power while becoming more cost-effective.
Buried in the storage graveyard are the remains of magnetic-core memory, the 3.5-inch floppy disk, CDs, ZIP drives, and DVDs. Are HDDs next? Maybe not in the short term, but SSDs are pushing traditional hard drives closer to the trash bin.
About the author
Ty MacAdam is the Product Marketing Manager for the next-generation Data Center SATA Solid State Drives in Intel’s Non-Volatile Memory Solutions Group (NSG). He is responsible for the strategic planning, product marketing, and go-to-market strategy for Intel SSDs.