Reports that hamr has been delivered to our cloud costumers. ! So will this be are saviour?
18 replies (most recent on top)
I used to do drive qualification for a US computer maker.
Basically, our process was to put all the drives we were qualifying into systems, then put those systems inside an environmental chamber. We'd then start running an extremely I/O intensive workload. While the workload was running, we would adjust the temperature in the chamber from 50 F to 95 F, while simultaneously adjusting the voltage to the system power supplies from 90 VAC to 130 VAC. We would expect the storage subsystem to operate at all four temperature/voltage corners.
We'd also put the drives into systems and drop-test the systems from a variety of heights and orientations. The drives (and systems) needed to keep working after a variety of drops.
Finally, the systems would be expected to work while on a vibration table, although we didn't skew voltage and temperature while on the table (our chambers weren't big enough for the tables).
Another team did performance testing of drives being qualified. That was outside my area of responsibility.
This was years ago, likely things have changed somewhat since then, but I'm sure the idea is still the same. The drive has to operate error-free at expected performance levels for every possible combination of temperature, voltage, and shock that we warranted our systems would work at.
In the days of PATA, every combination of drive makers on the same cable also needed to work at every corner. We'd expect our factory to connect drives without manufacture restrictions, so a Quantum and a Maxtor needed to work together on the same cable without issue.
Didn't STX always say we have a technology lead over the competitor, but missed out that the competitor have higher market share than us?
Seagate qualifies our drives, then customers qualify the drivers for their purposes.
Customers often do not tell Seagate what they really need, or provide any test data.
But it is standard practice to test products in-house.
data center customers won't risk reliability until HAMR demonstrates reliability and TCO comparable to existing technologies, so there's no fast ramp to revenue
Cust (not Dropbox), has gotten units to test in their systems. After a few tests (months if not quarters), they will begin paying for them and hopefully ramping their demand up. Other products being nixed to get to HAMR faster because the jump in capacity per disc is so far beyond what the competition will be able to get to for years. So the faster we qual and ramp HAMR at multiple customers the better for our revenue and share.
So Seagate gives a some HAMR drives to Dropbox and then uses the corporate jet to shuttle their CEO around to gain favor. This doesn't seem like a way to rescue the company from their cash crisis - not paying suppliers, putting the headquarters up for sale, etc, etc. HAMR revenue is still years away. The company probably in bankruptcy before then. This is why Nygaard is leaving. Who's next?
Nope. Seagate leaders maxed out it's debt and so has no ability to raise capital to ramp HAMR other than extremely expensive junk bonds. Obviously cash flowing it looks to be a bit of a problem right now. Death spiral has begun. I'm hoping it's purchased by another company to run as a legacy business before it's sold off for scrap. I still remember many all hands meetings being lectured that the huge debt was no problem at all and anybody who thought so was just ignorant about financials.
We need to ship more CMR drives right now. That's the only way we are gonna be safe.
the post says "costumer" and not "customer".
Very honest: Cost is high > cost -umer.
How about "cost-hamr"
How much does the HAMR drive cost?
Only if HAMR drives are not sold to CCP affiliates like Huawei
HAMR works, it's whether customers find that the reliability and cost make adoption worthwhile.
How many billions are needed in capital to scale HAMR? If a cloud service provider wanted to take the risk.
@OP+1miaEBSH doubtful because adoption and meaningful revenue are years away
Sadly we can’t sell our way out of this problem even if we had a market willing to do that. So raising massive amounts of cash as fast as possible is the only option. Lay offs won’t be enough by a long shot. So grab some popcorn sit back and enjoy the show!
Going by these comments I'll still take it - that we don't know if hamr really works... 👍
What does "qualified" even mean in this case? Whose qual: Seagate or customer? Not clear.
@set+1miaEBSH the articles says they were qualified hamr drives shipped / so do you think seagate is still clutching at straws with hamr ?
Didn't they announce at the call that they delivered samples for evaluation? Not an order for drives. They have been sending samples out for years.