I have heard about HAMR for about 20 years, yet it is still not successfully commercialized.
I am wondering if HAMR will eventually be successfully? If not, it will be a good lose to the competition to WD, since it's a bad bet, a big strategic bet.
27 replies (most recent on top)
Hammer will not work like normal drive. It reliability will be bad compare to all past generations especial in write mode. behave more like CD where writes are limited.
: @khvu+1smDzDDP And IOPs is an issue for drives, so it will be a trade off
DM is a liar
SSD runs hotter especially NVMe, good luck.
: @jfdx+1smDzDDP cost was used to debate whether flash would supplant drives in PC and that has happened, so at some future point a similar crossover will happy with data center drives
Flash NAND tech greatest disadvantage besides being more expensive is a limitation on the number of writes, this technology does not clear or erase it has to overwrite data. So, even if you plan to use an SSD for 5 years you may end up using it for only a year or two if you write to it more frequently. Good luck.
Pure Storage is a contender and we as an industry too often dismiss them
can infer from public comments about the strength of their roadmap that one of our competitors may be lagging after they detoured a few years back to chase another technology
Seems like Toshiba has figured out HAMR since Toshiba achieved 32TB on 10 disks. Amazing that they were able to do it with HDWY heads and Resonac media with HAMR SMR.
https://www.storagenewsletter.com/2024/05/20/idema-symposium-toshiba-demos-nearline-3-5-inch-hdd-with-32tb-on-10-disks-with-hamr/
Toshiba plans to start shipping test sample HDDs with HAMR technology in 2025. Pretty sure Seagate has shipped many HDDs samples already with CMR at these densities. It seems like Seagate could be 2-3 years ahead of the competition.
No comment on life which is the main issue for all these drives...
Still no announcement or data from WDC on HAMR...
My guess, it is about a year out for all kinks to be fixed to an acceptable level, or stable. Then, mfg will start making them like crazy! The pendulum will swing back, SSD too expensive still to be practical.
a bottle rocket
It’s coming fast like a rocket
@7zgy+1smDzDDP same bold predictions for yrs, so where is it
HAMR will be most successful product of this decade
capacity demand accelerating SSD adoption and HAMR is likely going to be too late to fully intercept or meaningfully disrupt this transition
HAMR has been a want a be product for 10 years now. Will never be successful. Focus on SSD tech or just say goodbye to it and move on.
Just like I said a month or so back, this company is following the failed Boeing strategy of off loading quality and cutting back on long established documentation practices. That's what you get when you allow shareholders to run the company.
They got rid of supplier quality engineering almost entirely so suppliers are policing themselves and making decisions about quality with no oversight really. Supplier audits, corrective actions, improvement plans, SPC charting, lot acceptance testing guidance, etc. There's just a lot of empty chairs now at Seagate to do that stuff now. Well, this is what you get.
To make it work, the people who actually could make this work are all long gone.
So many hick-ups STILL after all these years. Big D had to answer to the analysts and excuse it with “component” issue (true) - but something this should have been caught way way earlier. So embarrassing.
Now everyone is scared to death, make everything (even more) extra clean, remove risk going single source, closely timed builds, etc.
Maturity my a-s. What a clown show.
unlikely the TCO makes it more than a niche product and by then flash will occupy more of the data center that matters
No
Digigate
Seagate Digital or Westgate?
Flash has essentially replaced drives in consumer products, relegating them to security and surveillance market, and flash will continue to displace high-performance drives, leaving drives a slice of the data center pie.
HAMR drives have been tested by customers for a long time, and we're still waiting for adoption. Using HAMR involves increased complexity in the drive, which not only drives costs up, but impacts reliability and negatively impacts yield. The real problem is that we can't boost drive capacity using the older technologies anymore: HAMR has to work AND be adopted for Seagate not to stagnate. Failure would mean that our customers will start to wonder if they shouldn't simply make the jump and depend on SSD for their future projects.
HAMR is all there is now to showcase a next generation technology, but we know meaningful revenue and margin contribution will take time. SSDs like the recently announced Samsung 100TB (V9) will continue to erode high performance drive share, a market WDC all but abandoned when they acquired SanDisk.
Depends on definition of success. Customer shipments are announced and public. I think they need to test it on their systems in volume before they fully commit but I know nothing. I think there is only going to be room for 1 legacy HDD maker in the near future. Pretty much guaranteed there will be a WDC and Seagate merger in coming years. It will happen when SSD starts to steal significant cloud market share which is not too far away. I mean look at the revenue trends. The reduction in redundancy with a merger and better pricing power will be appealing. No fear of government block of SSD is still a competition