Let's talk about the (real) price of flash and spinning disks

The winners and losers in a battle between NAND and revolving rust

Comment Lately I have been writing a lot about the role of flash memory and disk drives in the future of storage infrastructure (here and here are a couple of examples). And it's clear flash will be used for primary workloads, and object storage and disks for the rest.

The reality is that flash memory is quickly becoming already a no-brainer when it comes to IOPS per GB in primary data scenarios. But where there is a demand for capacity, disks will continue to fulfill that appetite for a long time yet. Even if you find a good deal on solid-state storage, and flash prices continue to shrink, the difference between that and disk in terms of $/GB will still be huge over the next few years.

Your mileage may vary

$/GB is or was the parameter used in the past to calculate the cost of storage with traditional arrays. But things have become a lot fuzzier with modern data services and compression techniques. $/GB could vary substantially from user to user, and it often depends on the type of data you are storing, its compressibility, and other factors related to the overall efficiency of the storage system and how it is used.

Last week SanDisk launched a new box, a sort of “JBOF” array (Just a Bunch Of Flash). It’s amazingly dense and cheap (less than $2 per GB, they say) but if you add data protection (a RAID or a file system, like Ceph or ZFS for example), things could change. Just think about mirroring it, the $2/GB price tag suddenly becomes $4 per GB.

Reality check

Just to make it easier for everyone, I’m going to reveal facts about $/GB and $/IOPS in the easiest possible way. I'll compare prices found in a recent HGST price list. These are list prices for a single disk purchase (which means that the final price could be a lot less if you are an OEM or you buy in quantity). One of the prices comes from Amazon, and this is a consumer-grade product.

Here are the four disks:

  • Performance-focused:
    • HGST Ultrastar C10K1800: 2.5” 10K RPM, 900GB SAS disk – $350
    • HGST S841: 2.5” MLC 15nm 800GB SAS SSD – $2,440
  • Capacity-focused:
    • HGST Ultrastar He8: 3.5” 7.2K RPM 6TB SATA disk – $298
    • HGST Deskstar: 3.5” 5.4K RPM 4TB SATA disk – $150 on Amazon.com

First of all, we are only comparing simple components (raw material), and any operation you perform on the front-end of an array produces several operations in the backend. RAID, block-size optimization, compression/recompression, deduplication, bus or CPU limits, and so on, limit the number of array IOPS effectively seen by the hosts.

Let’s do the math:

Table comparing capacities, prices and IOPS

Considering a 50,000 IOPS 4KB random workload (which won't be the IOPS you obtain from your array)

It’s plain that the solid-state drive wins hands down if the comparison is about $/IOPS and IOPS/GB. But on the other hand, there is a great abyss between the 0.037-0.049 $/GB registered by the SATA drives and the others.

And the loser is ...

It’s also quite clear to me that the big loser here is the 10K SAS disk. It doesn’t come out the winner in any comparison, and the technology won’t be developing that much in the future; results can only get worse over time.

At the same time, the consumer-grade SATA disk's $/GB is almost 100 times better than flash. Yes, 100. To be fair, a consumer-grade flash drive (with SATA interface) is only 10x the $/GB of a SATA hard disk, but even with a year-on-year price drop of 40 per cent, it will take five to six years to have comparable prices, without considering that we already have much higher capacity hard disks in the pipeline.

In the near future we will be seeing many more (petabyte-scale) storage systems using consumer-grade, low-cost SATA drives, coupled with modern data protection techniques like distributed Erasure Coding, for example. This helps to mitigate multiple disk failures and longer rebuild times. For example, HDS, with its HCP S10, is already doing this.

Closing the circle (and stating the obvious)

Flash wins in IOPS, hard disks win in capacity, and this won’t be changing anytime soon.

All-flash storage systems are cost-effective because flash enables the use of in-line deduplication and compression at virtually no cost. Their predictability is very high, and they are good for highly demanding primary workloads.

At the same time, hybrid systems are better than all-flash when the demand for IOPS is not too high. In this case, the IOPS/GB shown in the above table is an important metric. If you want to build a balanced system in terms of IOPS and capacity, you still need flash and SATA disks. This is why most SMB customers are still going hybrid, and companies like Tegile and Nimble are doing so well.

Last but not least, high-capacity systems will continue to be nearly all-disk for a long time. I say "nearly" because most object storage systems today use RAM for metadata and cache. Tomorrow, with very high capacity disks, it could be cheaper to store some of the information on relatively small solid-state drives. ®

Similar topics

Similar topics

Similar topics

TIP US OFF

Send us news


Other stories you might like