Seagate To Double HDD Speed With Multi-Actuator Technology

We already know that HDDs suddenly have a new lease on life with new recording technologies, but now your next hard drive may be twice as fast as any drive on the market. Seagate announced that it plans to double the performance of its future HDDs with its multi-actuator technology.

HDD density has continued to rise, but the mild performance improvements with each new generation are mostly borne of density increases instead of new whiz-bang tech. That's why SSDs have taken over the performance role and HDDs have retreated into the high-capacity space. We're on the cusp of even more explosive capacity increases as WD and Seagate both roll out their new recording technologies. Both companies predict their future HDDs will increase beyond 40TB, which means stagnant performance will become one of the biggest challenges.

Increased capacities combined with similar gen-on-gen performance results in a lower IOPS per TB rating, which isn't ideal for data centers. For instance, most 3.5" enterprise HDDs reach ~250MB/s of sequential throughput, so a 40TB HDD would take nearly two days to fill under ideal conditions. That also has implications for application performance, RAID rebuilds, and other common tasks.

How Seagate's Multi-Actuator Technology Works

Seagate's multi-actuator technology is a simple concept, and the idea certainly isn't new. In fact, the company has already developed drives with multiple actuators in the past, but they weren't economically viable due to higher component costs.

Most HDDs read and write data to and from multiple platters. For instance, Seagate's largest drives wield up to 8 platters and 16 heads. The heads, which are connected to the end of an actuator arm assembly, read and write data from both sides of each platter.

Unfortunately, those 16 heads are all aligned on the same arm, which means they all move in unison. Simultaneously aligning all the heads on all the platters isn't possible because of the incredibly thin data tracks on the platters, so only one of the heads is actively reading or writing data at any given time. That limits read/write throughput and performance with randomly accessed data.

Seagate's new design uses two sets of actuator arms that operate independently. Each carries eight heads. That means the drive can read or write from two heads at once, provided they are attached to different actuator arms. The drive can respond to two commands in parallel and one head can read while another writes, or both can read or write simultaneously.

Seagate's animation appears to show an additional actuator magnet (the far left of the image) that manipulates the VCM (Voice Coil Magnet) assembly that moves the arm to position the heads. The additional actuator magnet will likely add at least some cost to the equation. However, both actuator arm assemblies operate on the same pivot point, which is more cost-effective than using an entirely separate assembly. That's one of the keys to avoiding the cost challenges the company encountered with its early multi-actuator models. The single pivot also helps reduce redundant components.

The heads read data from servo tracks embedded onto the platter, which in turn allows the actuator magnets to adjust head positioning dynamically. In current models, Seagate also has a PZT micro-actuator near each head that provides additional fine-grained positioning. Managing head positioning, data reading and writing, error correction, and other tasks require a dedicated processor (controller) onboard the drive. In the past, the additional processing overhead from managing a separate set of heads required an additional controller, which adds cost, power consumption, and complexity. Today, it's likely that a single modern drive controller can handle the task, albeit with a slightly more powerful model that might have higher power consumption.

The additional actuator magnet/VCM assemblies will likely increase drive weight and power consumption to some extent. Seagate apparently feels it is a manageable overhead and that overall lower dollar-per-TB metrics will offset any increases in manufacturing costs. 

But Will It Work Like A Normal Drive?

Seagate has not confirmed that the new multi-actuator drives will adhere to the standard dimensions of the 3.5" form factor, but that seems likely, because it would ensure broad compatibility with existing infrastructure.

Seagate's blog post describes the process as:

The host computer can treat a single Dual Actuator drive as if it were two separate drives. This means the host computer can ask a single high-capacity drive to retrieve two different data requests simultaneously — delivering data up to twice as fast compared with a single-actuator drive.

We followed up for further clarification. The company responded;

The device shows up as one worldwide name to the user with two access streams available for communication. The drive is one volume but split into two spaces that the user can communicate to. And in the future, if a quad-actuator were implemented (for example), that would be one worldwide name to the user, with four access streams available for communication.

Seagate says the drives could use SAS, SATA, or NVMe interfaces, but the company will respond to the needs of its customers to develop the final solutions.

For instance, the venerable SAS interface features dual-port connectivity that provides two separate pathways into a single drive. The additional pathway can be used for multi-pathing or fail-over. The pathways can also be combined into a dual port connection to offer up to twice the performance from a single drive, as we see with some of Seagate's cutting-edge enterprise SSDs. The NVMe protocol also supports similar features.

Seagate will share further technical details as specific models come to market.

Thoughts

As we've seen unfold quite unexpectedly this year, the humble HDD suddenly has a new path forward. Seagate is moving forward with HAMR (Heat Assisted Magnetic Recording) technology that uses tiny lasers to increase storage density. The company plans 20TB+ drives in 2019 and 40TB+ models by 2023.

But WD isn't standing still. WD plans to use MAMR (Microwave Assisted Magnetic Recording), which the company says can enable up to 40TB HDDs by 2025.

The change to two different recording technologies marks a tremendous shift; the industry has largely used PMR (Perpendicular Magnetic Recording) for more than a decade.

The two new technologies will compete fiercely in the market. Seagate's Multi-Actuator technology may give it an advantage over competing drives from WD and Toshiba, but much of that hinges on the ability to mass-produce the drives cost-effectively. Speed is nice, but HDDs are all about cheap and dense storage, so pricing will be the key to success for Seagate. Adding moving components also brings the chance for higher failure rates, but Seagate's design appears to remove as many redundancies as possible. We imagine the company will strive to deliver the new drives with the same reliability ratings as current models, but we won't know more until the drives come to market. We still aren't sure of Toshiba's plans or if WD has its own performance-boosting technology in the works.

SSDs may have replaced hard drives for many performance-sensitive workloads, but they will continue to be relevant in bulk storage roles for many years to come. Increasing HDD performance won't place them on par with SSDs, or even close, but it will address the IOPS per TB challenges that will crop up with next-gen recording technologies.

Seagate's plans to pair HAMR technology with multi-actuator technology to eventually provide four times the capacity and twice the performance of existing hard drives is promising. It certainly looks like the old HDD has plenty of gas left in the tank.

Paul Alcorn
Managing Editor: News and Emerging Tech

Paul Alcorn is the Managing Editor: News and Emerging Tech for Tom's Hardware US. He also writes news and reviews on CPUs, storage, and enterprise hardware.

  • deepblue08
    Surprised this hasn't been done before. So the performance should be comparable to that of 2 HDDs in RAID 0?
    Reply
  • PaulAlcorn
    20496544 said:
    Surprised this hasn't been done before. So the performance should be comparable to that of 2 HDDs in RAID 0?

    Performance should be on par with a RAID 0 of comparable drives.
    Reply
  • King_V
    20496544 said:
    Surprised this hasn't been done before. So the performance should be comparable to that of 2 HDDs in RAID 0?

    Agreed - for several years, I'd wondered why there hadn't been drives developed, particularly in multi-platter scenarios, with separate actuators, and maybe the possibility of built-in RAID-like functionality.
    Reply
  • vern72
    It's just a matter of time before each pair of heads get their own actuator.
    Reply
  • Geef
    It might be a negative thing if you ever Defrag one of these hard drives since it might stick all the files on one area and the extra speed wouldn't be used.
    Reply
  • mrmez
    "That means the drive can read or write from two heads at once, provided they are attached to different actuator arms."

    So it's like RAID 0, but not all the time.
    Reply
  • chicofehr
    I'll wait to see how reliable they are first but its a good start. SSDs are still too expensive for storage purposes.
    Reply
  • alextheblue
    20497701 said:
    "That means the drive can read or write from two heads at once, provided they are attached to different actuator arms."

    So it's like RAID 0, but not all the time.
    If you stripe the data, you get a RAID 0 array all the time. What I want to know is why isn't this done in hardware at a drive level, invisible to the OS?
    Reply
  • cryoburner
    but now your next hard drive may be twice as fast as any drive on the market.

    Maybe... If you buy the highest capacity drives and don't plan on buying one for a while. I get the impression that this is most likely to only appear in the highest capacity models, at least initially. The cost of adding these components to a drive priced in the sub-$100 range might be notable, but with the highest capacity 8-platter drives priced around $500, the cost of doubling up these components should make a relatively minimal difference to the total cost of such a drive.
    Reply
  • elbert
    Increasing the read/write speed is fine but can not compete with SSD on that front. Given the rising cost of SSD and memory as of late HD makers seem ok with throwing a golden opportunity. They just go along with extremely outdated decade+ capacity for $50 in 1TB. To make matters worse they also sale right at decade old capacitys at $75 with their 2TB. The 4TB drive capacity is turning 8 years old in 2018 and its anywhere from $120~200. Point is people already have them and forgo a small SSD with hard drive to pick up just a larger SSD. Last few years HD makers have fallen to point that SSD are of the capacity of their low end models. This is in part the reason memory/SSD prices are at all time high.
    Reply