Seagate HAMRs out a roadmap for future hard drive recording tech

100TB by 2025/26? Bring it on

Seagate has set a course to deliver a 48TB disk drive in 2023 using its HAMR (heat-assisted magnetic recording) technology, doubling areal density every 30 months, meaning 100TB could be possible by 2025/26.

HAMR overcomes the tendency of smaller magnetised areas in current PMR (perpendicular magnetic recording) technology to flip their magnetic polarity, and hence their binary bit value, through temperature changes and interference from neighbouring bits.

A different and more stable recording medium is used, which needs heating before bit values can be written, meaning the read-write head requires a laser-heating element. HAMR bits can be made smaller than PMR bits, so a disk drive platter's areal density increases and drive capacity rises beyond the 16TB limit or so of 3.5-inch format disk drives.

Western Digital is developing alternative microwave-assisted magnetic recording technology (MAMR) to overcome the PMR issue.

Seagate will introduce its first HAMR drives in 2020.

Seagate_Why_HAMR_950

Click to enlarge

The chart above, from an A3 Tech Live event in London, shows Seagate started developing its HAMR tech in 2016 and that a 20TB+ drive will be rolled out in 2020.

The last PMR drive appears in 2019/20 with 16TB capacity. Seagate's current highest-capacity drive is a 14TB Exos 3.5-inch product.

There is a forecast of areal density doubling every 2.5 years, and Seagate shows two other HAMR drive capacity points: 36TB in 2021/22 and 48TB in 2023/24.

Capacity goes on increasing beyond 2025, with 100TB looking likely.

The firm makes the point that HAMR drives will be drop-in replacements for current PMR drives.

Seagate will actually develop performance-optimised HAMR drives with MACH.2 multi-actuator technology – two read/write heads per platter – and capacity-optimised drives with shingled magnetic recording (SMR). These are shown in a second chart:

Seagate_HAMR_roadmap_650

Click to enlarge

SMR drives have blocks of partially overlapping write tracks exposing narrower read tracks, making read speed faster than re-write speed as the overlapping track block has to be rewritten in its entirety rather than, as at present, just the affected bytes in a write track.

The two HAMR drive product lines have approximately the same capacities, but the performance drives appear to reach a particular capacity point a little after the capacity drives. This could be an artefact of the graphic and, anyway, Seagate is not predicting product announcements here, only capacity trends.

The first multi-actuator drive should appear in 2019, according to the chart. That could potentially speed up SMT drives as well as non-shingled disks.

When HAMR tech loses its ability to advance areal density further, Seagate envisages using HDMR (heated dot magnetic recording), a refinement of HAMR that uses BPMR (bit-patterned media recording) in which smaller bits (made up of individual magnetic grains) sit on raised dots or islands a few nanometres above a baseline. That would effectively increase the separation between neighbouring bits.

We have 13 to 25 months to wait for the first HAMR drives. Seagate said it has made millions of HAMR drive heads and the technology is reliable. Bring it on, we want to see the HAMR drop and deliver both fatter and faster drives. ®




Biting the hand that feeds IT © 1998–2018