So, if an enterprise buys ten 200GB SSDs it could then choose not to buy fifty 200GB HDDs.
and why would they want to buy 50 x 200GB hdd in this day and age? Does anyone still make such low-capacity hdd anyway?
WD is experiencing global disk drive cooling as the slowing PC jetstream and flash El Nino jointly disrupt the HDD environment. But that's enough in the meteorological metaphor line. WD revenues fell 21 per cent year-on-year to $2.8bn in WD’s third fiscal 2016 quarter, which closed April 1. The quarter-on-quarter compare …
Clients don't need much.
My standard image is less than 50Gb. Even with 100Gb of roaming profiles, they're unlikely to hit 200Gb at all in a four-year PC lifetime (and they get re-imaged once a year or so).
I'm much more likely to buy old stock like 50 200Gb hard drives than even a handful of 1Tb hard drives. I'll get them in the box, untouched, brand-new and able to service 50 clients in one purchase because they are quite literally throw-away costs.
Servers, storage arrays, etc. are another matter. Windows 7 / 8.1 / 10 clients? Barely need 200Gb. Even the cheapest new machines nowadays come with 500Gb, usually, and that's overkill. One of the reasons that I'd rather buy a load of 256Gb SSDs than 1Tb hard drives for them all. More than adequate on size, at a decent price point, and a user-visible "upgrade" in terms of speed.
"Enterprise SAS drives still come in such sizes."
But noone's buying enterprise SAS spinners anymore(*), because SSDs blow them away in IOPS.
(*) Not in any numbers which matter. Enterprise SAS "performance" hdd sales have fallen off a cliff with only "bulk" spinners still selling in any numbers. The days of 15 and 20krpm drives are over.
Because they are shooting for IOPS and not maximal space. Even fast hard drives may only get 500 IOPS. A single SSD can easily provide 50,000 IOPS.
But beyond your question, everything in enterprise storage is about reducing storage utilization. Most SANs offer deduplication and compression. Coupled with an SSD tier, many businesses realize they need less total storage with newer technologies yet still have very high performance.
There is still need for HDD in some corporations. Some, such as Google use them for (seriously) large data storage resulting from cloud content. Don't let the title fool you but check out:
And what's more, they want more platters per HDD for failover reasons. Strategically, there is still some validity in HDD... But not for home use, I would not go back to HDD - SSD is the best path forward (at this point until something better comes along).
Manufacturers cheating and hiding money... That is what is really happening.
People buy more HDUs and SSD now than ever before. And faster internet connections in major countries means higher demand.
The enterprise sector needs more drives and data centers for cloud services or not need to buy more units.
So these manufacturers now claiming to be losing money and no one buying drives anymore are just liars and telling b*ll.
Am I doing the math right?
It appears that in the past you purchased enterprise disks to achieve high IOPS: 10K and 15K RPMs with SAS. But these disks can still usually achieve less than one IO per revolution when tranactions are truly random access, and 15K RPM is only 250 RPS, so for databases we were IOPS-limited, not size-limited, and we spread a database across lots of disks. With an SSD, we are limited (to a first approximation) by the SAS, at 12 Gbit/sec or about 1.2 GB/sec (counting 20% overhead) so for reasonable IO sizes we get more than 50,000 IOPS. It looks to me like a single 1 TB SSD would displace more than 200 enterprise HDDs.
Obviously, if I also need very large files, I am in a different situation, but these match with slower HDDs where I can transfer entire tracks at once.
"So, if an enterprise buys ten 200GB SSDs it could then choose not to buy fifty 200GB HDDs. "
More to the point, they're highly unlikely to buy those SSDs from Seagate OR WD or their many tentacles.
WD and Seagate are circling the drain. They may buy up some of the smaller SSD names, but they can't rely on OEM sales volume anymore, consumers have taken note of HDD unreliability (plus the brand names involved) and anyone except "buy the cheapest possible" will be looking for 5+ year warranties.
It may well end up that WD ends up as a brandname for Toshiba and Seagate for Samsung, but the accountants would have to balance brand name recognition against the well-poisoning that these two companies have indulged in over the years since the Thai floods.