Seagate: 30TB second-generation HAMR hard disks are almost here

[Alfonso Maruccia]

Forward-looking: Seagate has invested a huge amount of money and energy into a HAMR-powered future for storage based on magnetic technology. Now the company says it's almost ready to bring its next generation of HDDs to enterprise and consumer customers alike.

The future of magnetic storage will begin in the third quarter of 2023, as Seagate has announced its latest plans to start mass production of second-generation HAMR-based hard disks. In its latest conference call for investors, the storage giant said that the launch of its newest HDD "platform" will be "ahead of schedule," while efforts to improve yields (and, of course, margins) are ongoing.

Despite the dire situation the storage market is experiencing right now, Seagate is looking at the future: Heat-Assisted Magnetic Recording (HAMR) uses a laser diode attached to each recording head to "momentarily heats a tiny spot on the disk," the company explains, so that data bits written on a magnetic platter become smaller and more densely packed than ever "while remaining magnetically and thermally stable."

Seagate has already shipped limited amounts of HAMR-powered HDDs to selected customers for testing, but the technology is seemingly almost ready for the mass production of consumer and enterprise hard drives. The speed of the initial HAMR volume ramp-up "will depend on a number of factors," Seagate said, such as product yields or customer qualification timelines.

For the first and second quarters of 2023, Seagate still plans to rely on conventional magnetic recording technologies to push HDD capacities to new heights. In Q1 or Q2, the company will release new 10-platter models using perpendicular or conventional magnetic recording (CMR/PMR), cramming 2.2 terabytes on every platter for a total 22TB capacity. Using the slower shingled magnetic recording (SMR) tech, Seagate will offer 24TB models to customers willing to adopt it for "certain applications."

Second-gen HAMR platters will eventually be able to cram 5 terabytes of data on a single platter, Seagate estimates, while the first mass-produced HAMR HDDs will offer "only" 30 terabytes of nominal storage capacity. HAMR hard disks with 40TB of storage are expected for 2024 or 2025, while 50TB models could arrive in 2026.

For now, the main problem of HAMR drives is yield – and it will stay like this for the upcoming years as well. In 2023, Seagate said, the share of HAMR-based HDDs on the market will probably still be "relatively low." Major production improvements should arrive in 2024, with further acceleration in 2025. Improved yields will bring the heat-assisted technology to midrange or even low-range HDD models as well.

Permalink to story.

https://www.techspot.com/news/97409-seagate-30tb-second-generation-hamr-hard-disks-almost.html

 

[VitalyT]

And "almost" irrelevant.

It's like working on a new car with V12 in today's market, both admirable and dumb.

 

[rmcrys]

A product that makes a lot of sense for companies that need loads of space / backups and maybe to replace all consumer 3.5" disks with high capacity 2.5" disks (NAS).

Most consumer NAS run on 1G or 2.5G Ethernet which limits the transfer to 100 - 280 MBps (at best) for all hdds. So if I use HDDs it will be fine, an SSD just brings silence because even a sata basic one will go up to 450 MBps.

Other than that, they should invest on a cheap slow high capacity SSD (like one up to 0.5 to 1 GBps but with 5 TB for under 200€)

 

[Plutoisaplanet]

And "almost" irrelevant.

It's like working on a new car with V12 in today's market, both admirable and dumb.

I wouldn't exactly characterize it that way. I'd expect there to be excellent uses to hard disk drives for decades to come for commercial reasons as opposed to consumer reasons. As long as this is more cost-effective per TB, then there will always be a use for HDDs when the benefits of SSDs aren't as relevant. From that perspective, ongoing investments in this space definitely pays itself off.

It's more like investing in semi trucks or other work vehicles in the face of EVs becoming commonplace. The ROI of a new program for these use cases just aren't the same as targeting the mass market, and the needs are different enough that the old way of doing things will continue to be better for the foreseeable future.

 

[Avro Arrow]

The only question that must be asked about this technology is:

"Will it be fast enough, efficient enough or cheap enough to compete with SSDs?"

If the answer is "no", then Seagate is just flushing their R&D money down the toilet but if the answer is "yes" then this will be a really interesting technology.

 

[Kashim]

And "almost" irrelevant.

It's like working on a new car with V12 in today's market, both admirable and dumb.

Sorry, but it is your comment on this technology that is irrelevant. Do you really think that a company like Seagate and other multi-billion dollar companies would invest so heavily in such a technology if it was irrelevant?

Data centers, backup servers, etc. are still using mechanical drives and will be doing so for a very long time. Mechanical drives are still far far cheaper per terabyte compared to solid state drives. People like you who only need a few TB of fast storage to quickly load whatever crappy game you happen to be playing obviously won't care about mechanical drives. But of course you won't even consider other applications for such a technology because... why would you? If you have no use personally for something then obviously it's irrelevant to everyone in the world. *roll eyes*

 

[DaveBG]

Awesome finally some improvement

 

[Kwesley]

Sorry, but it is your comment on this technology that is irrelevant. Do you really think that a company like Seagate and other multi-billion dollar companies would invest so heavily in such a technology if it was irrelevant?

Data centers, backup servers, etc. are still using mechanical drives and will be doing so for a very long time. Mechanical drives are still far far cheaper per terabyte compared to solid state drives. People like you who only need a few TB of fast storage to quickly load whatever crappy game you happen to be playing obviously won't care about mechanical drives. But of course you won't even consider other applications for such a technology because... why would you? If you have no use personally for something then obviously it's irrelevant to everyone in the world. *roll eyes*

I agree with everything you said, but tone it down brother lol. You’re acting like the guy tried taking your girlfriend away

 

[PEnnn]

Sounds like a DVD burner technology, but way more advanced.

Might convince some people to still use HDDs (especially data centers, but also some consumers), assuming the price is right.

SSDs might be much cheaper by then and still with no moving parts.

 

[Marco Mint]

The only question that must be asked about this technology is:

"Will it be fast enough, efficient enough or cheap enough to compete with SSDs?"

If the answer is "no", then Seagate is just flushing their R&D money down the toilet but if the answer is "yes" then this will be a really interesting technology.

I'm sure Seagate are more aware of the technology and its possible place being viable within the market than someone who knows little about it, or the market they are targeting within their own company. These decisions aren't just thrown up in the air and gone with on gut feeling.
A 4TB nvme drive is approx the same price as a 20TB conventional HDD currently. If you want/need LOTS of space for lots of video files etc, the 20TB is going to be the better option. You are thinking within the narrow gap of your own personal needs and not taking into account anybody elses possible needs.

 

[Hodor]

Seagate can name the disks Mc HAMR.

 

[pcnthuziast]

Good for those with large collections of uncompressed 4K HDR Blu Rays.

 

[toooooot]

The only question that must be asked about this technology is:

"Will it be fast enough, efficient enough or cheap enough to compete with SSDs?"

If the answer is "no", then Seagate is just flushing their R&D money down the toilet but if the answer is "yes" then this will be a really interesting technology.

I would also take "reliable"
7y warranty drive would be on my purchase list.

 

[rmcrys]

Good for those with large collections of uncompressed 4K HDR Blu Rays.

What you mean is no recompressed blu-ray, because they are already compressed.

This tech of for all those that need massive amounts of storage, most of them are companies like Netflix, Disney, Hollywood studios, etc.. that need thousands of these. Of course consumer hdds are losing huge amounts of market, even for those with a lot of blu-ray backups. Most blu-rays have up to 60 GB, so if you have 200 uncompressed, you will be more than fine with a NAS 2x 8 TB or, found even stronger with space and backup, there are several 4 bay NAS and 12 TB disks are very ok in price, so in 2023 you'll have plenty of space already.

 

[Mr Majestyk]

And "almost" irrelevant.

It's like working on a new car with V12 in today's market, both admirable and dumb.

Sure, provide us with 30TB affordable SSD's and problem solved. HDD's will disappear in a heartbeat. Oh wait, what, that will never happen. By the time we have 50TB HDD's, 8TB SSD's will still be a ripoff.

 

[Darth Shiv]

And "almost" irrelevant.

It's like working on a new car with V12 in today's market, both admirable and dumb.

30TB is not comparable. It's just not a huge market segment for demand but in certain cases like datacenters, the need for dense storage is absolutely there. I would say it's more analogous to a large train using older tech for motor compared to say a super fast small consumer vehicle. The train absolutely has a reason to exist and business case but most of the market has no such need for such an expensive vehicle or the capacity associated.

 

[Darth Shiv]

The only question that must be asked about this technology is:

"Will it be fast enough, efficient enough or cheap enough to compete with SSDs?"

If the answer is "no", then Seagate is just flushing their R&D money down the toilet but if the answer is "yes" then this will be a really interesting technology.

None of those metrics are relevant to this market segment. The question you DIDN'T ask is does it out-compete with SSDs on capacity/storage density? Look at bang for buck for 30TB of SSD. It's poor. A HDD clearly has no ability to compete with SSDs on speed.

HDDs *are* cheap enough on capacity per $. They absolutely have worth there. Same with capacity vs physical space. So from certain metrics yes actually "cheap enough" is achieved...

Not sure what you mean by efficient.

 

[Avro Arrow]

I'm sure Seagate are more aware of the technology and its possible place being viable within the market than someone who knows little about it, or the market they are targeting within their own company. These decisions aren't just thrown up in the air and gone with on gut feeling.
A 4TB nvme drive is approx the same price as a 20TB conventional HDD currently. If you want/need LOTS of space for lots of video files etc, the 20TB is going to be the better option. You are thinking within the narrow gap of your own personal needs and not taking into account anybody elses possible needs.

Corporations throw money at worthless tech all the time. Intel is the king of it and for every new technology that becomes viable, I'm willing to bet that there are at least ten that fail.

Quite often, they're just throwing things at the wall to see what sticks. Sometimes things stick for a bit but end up falling off. The problem with magnetic drives is that the media can only spin so fast before losing structural integrity and that spinning uses a lot more juice than an SSD.

If Seagate has found a more efficient read/write method, then this can be mitigated somewhat. In any case, I seriously doubt that it will supplant SSDs.

 

[rmcrys]

Corporations throw money at worthless tech all the time. [...] Quite often, they're just throwing things at the wall to see what sticks. Sometimes things stick for a bit but end up falling off. The problem with magnetic drives is that the media can only spin so fast before losing structural integrity [...] I seriously doubt that it will supplant SSDs.

Different markets here:
- consumers need relatively low amounts of space, they need speed. So SSDs are the better option and a couple of hdds as a backup
- enterprises or professional video workers need huge amounts of data space as most save their original video files. Here SSDs on the PC and hdds on their NAS will be the best option