Guest ArticleNews

Enter the Future: All-flash Data Centers are Now a Reality

Pratyush Khare, Vice President, Systems Engineering, Asia Pacific and Japan, Pure Storage
Pratyush Khare, Vice President, Systems Engineering, Asia Pacific and Japan, Pure Storage

In today’s fast-paced business environment, where data is growing exponentially and the global economy is shrinking, organizations must find ways to efficiently manage their data while reducing their carbon footprint. The data center represents a crucial area where businesses can improve economic efficiencies and sustainability. One technology that can help organizations achieve these goals is flash storage.

There is no denying the advantages of flash over legacy storage medium such as disk or tape, but the cost has been the prohibitive factor. However, the disk/flash TCO (total cost of ownership) cutover is closer than many people realize.

Let’s look at some of the key factors that are making flash the de-facto storage option for fast growing enterprises globally.

It’s important to note up front that flash has already been the obvious choice for any performance- or latency-sensitive workload for some time now. Since flash memory’s entry into the enterprise storage arena, it’s been knocking down hard disk tiers one by one, from top to bottom, until now there’s only one tier left where disk drives maintain a market share advantage: nearline, capacity-focused drives.

This is the lowest tier of online data storage, but it also represents a huge portion of bytes being stored, as well as the growing majority (more than 70%) of hard drive capacity shipped each year. That’s the tier that flash will penetrate next.

The price-per-bit for NAND flash is declining at a rate much faster than nearline hard drives. This, combined with the power, space, and cooling savings, higher performance, and better reliability of flash will soon result in hard drives becoming the less cost-effective option. Let’s go into a detailed look at some of these components.

Media Cost

The first-order cost is always going to be the media itself. While today the cost-per-bit of a hard drive is still lower than the densest flash, there are several key trends to pay attention to in 2023.

First, all the major tier 1 flash manufacturers are demonstrating significant density increases this year, over 200 layers of stacked 3D NAND in some cases. This increase in density will translate into better cost efficiency, as well as further improve power and space savings. Second, if analyst predictions hold, NAND prices overall are expected to decline through most of 2023, continuing a trend that began in late 2022.

Energy Cost

Data center infrastructure represents over 1% of global energy consumption, and this share is growing. As a result of rising energy costs, energy consumption by information technology has become not just a concern from an environmental standpoint but now also a significant economic one.

When looking at total cost of ownership, energy consumption will play an even more outsized role in 2023 and beyond, and infrastructure that is more power efficient will have an even greater economic advantage over the competition. As flash density continues to increase, power efficiency will surpass hard drives and thus contribute to a lower effective cost.

Higher Performance

In two keyways, the flash performance advantage over disk can translate into cost advantages.

First, because disk performance is still so low, even in workloads with modest performance requirements, you will often end up with stranded capacity. While spinning drives may hold 20TB, but if the performance of the system taps out at 16TB, you will not derive as effective a cost-per-bit. Flash doesn’t have this issue because read performance stays predictable even as drives fill up, unlike hard drives.

Second, any storage environment that focuses on resiliency will have a redundancy strategy which may involve storing multiple copies of data for example, in addition to keeping spare capacity around in the case of a failure.

This means that the performance of any one device is a variable in the equation used to determine how much redundancy you need to build into your system. Many years ago, hard drives had gotten large enough that after a drive failed, rebuilds would take so long that you needed to adopt dual parity strategies to ensure business resiliency. Today that can be triple parity or more.

With flash devices, faster rebuild times mean less bits need to be dedicated to resiliency structures, which equates to better effective cost efficiency. Some flash vendors offer robust, automatic resiliency tuning that takes all these aspects into consideration, without requiring users to determine what RAID level they need to set.

Better Reliability

Reliable devices need fewer replacements, and therefore lower costs. While most customers with support contracts don’t feel the cost directly, the cost of this unreliability is baked into the cost of those very support contracts. This is one reason why traditional storage vendors’ support costs would skyrocket in years 4, 5, and beyond.

To summarize, in almost every way, flash storage is superior to spinning magnetic storage, with the only major difference being cost. However, with the road to the disk/flash crossover point getting shorter and shorter, the dream of an all-flash data center will soon become a reality for most organizations. By embracing flash storage, businesses can improve their economic efficiencies and sustainability while meeting the ever-increasing demand for data management.

Related posts

o9 Promotes Anand Govind to Role of Chief Financial Officer

adminsmec

Avaamo Announces Reference Architecture for AI Applications, Powered by Intel Xeon Scalable Processors

adminsmec

Micron Ventures Fund II Commits $200M for Deep Tech Startups

adminsmec
x