Seagate recently introduced their 18TB flagship IronWolf Pro hard drive for SMB/SME NAS units. Today, the company is unveiling the Exos enterprise version of the same drive. The Exos X18 goes head-to-head against the Western Digital WD Gold 18TB EAMR-based drives introduced in July. Seagate is also taking the opportunity to expand their application/storage server lineup with a 2U 12-bay model - the Exos AP 2U12 Compute and Storage System.

The Seagate Exos X18, like the IronWolf Pro 18TB, is a 7200 RPM CMR (conventional magnetic recording) drive with a 256MB multi-segmented cache. It contains additional reliability features, and firmware tweaks to provide features geared towards data centers and enterprise storage arrays. These include caching tweaks to optimize the drive for low-latency large data transfers, and a power balance feature that allows customization of the power consumption for the best possible watts/TB given a particular set of workloads. The MTBF increases from 1.2M in the IronWolf Pro to 2.5M hours for the Exos X18.

The maximum sustained disk transfer rate also increases from 260 MBps to 270 MBps. Seagate quotes maximum operating power of 9.4W, with idling average at 5.3W. The IronWolf Pro 18TB idles at 5.2W.

Seagate also has a lineup of products under the Exos Systems tag - these include multi-bay rackmount storage enclosures falling under three different categories - the AP series for compute and storage with an in-built x86 server CPU, the X series RAID enclosures, and the E series JBOD enclosures. Today, the company is introducing a 2U12 AP model - as the name suggests, the 2U rackmount server supports up to 12 3.5" drives and comes with two 10GBASE-T ports and two gigabit ports. The server board is based on the Xeon v5 v4 family, and CPUs with TDP of up to 85W are supported.

Coming back to the Exos X18, Seagate offers it in both SATA and SAS versions. The SATA version has a MSRP of $562, undercutting the WD Gold by $30.

Source: Seagate

Comments Locked

38 Comments

View All Comments

  • Arbie - Friday, September 18, 2020 - link

    "Not to be a jerk"? With a nasty comment like that, on a trivial typo? Is this forum your bathroom?
  • ravib123 - Thursday, September 17, 2020 - link

    With regard to failure rates:
    I would say this is par for spinning hard drives. It’s a lower profit commoditized product, that is sensitive to drops and vibration. While I doubt they leave the facility bad then don’t make it to your door working quite often unless you purchase quantity.

    That first week of burn-in on a drive will determine if you’ve got a premature mortality on your hands.

    I see the same problems with all the brands, higher failure reporting from seagate and typically a better type of final failure, at least in our use case.
  • melgross - Thursday, September 17, 2020 - link

    I’ve never gotten a DOA drive in all the decades I’ve been buying drives. It was a surprise. To get two out of three wasn't encouraging. I imagine that there was a bad batch. But it just makes the whole thing unnerving. I’ve had drives work for a few months and die. Most last years, but I change them out after three years anyway.
  • leexgx - Thursday, September 17, 2020 - link

    This is why I norm split the purchases between company's (in your case 2 from one place and 2 from Another so can keep raid6 happy) sounds like you was going to be using RAID1

    But on that note 16TB disks I would strongly recommend using 4 disks in raid6+1 hotspare to auto rebuild in the event of a predicted failure or flat out disk fail as its going to take sometime to rebuild a 16TB disk (really any disk bigger then 1TB should be using RAID6)

    Raid is not a backup so you should have a similar setup just in case as it's a lot of data to go puff
  • ravib123 - Friday, September 18, 2020 - link

    ouch, I used to do that but the performance metrics between vendors for the pout poses if raid are quite different.
  • edwardhchan - Thursday, September 17, 2020 - link

    Ouch... I ordered 3 12tb Exos X16 drives and all 3 were fine from Newegg last month so there's that.

    and 16GB 7300RPM?? You are exhausted! LOL
  • Beaver M. - Friday, September 18, 2020 - link

    And here we thought Seagate HDDs would have become more reliable.
    Thank god I bought a 12 TB HGST, because I almost bought a Seagate one.
  • JKJK - Wednesday, September 23, 2020 - link

    I have 11 of the 16TB version, and they've been running fine for the last couple of months. Yes, it's too early to tell, but I hope they last.
  • JKJK - Wednesday, September 23, 2020 - link

    I don't have good experience with seagate from earlier, but it was that much cheaper that I had to try it. I once lost 16 seagate drives in one server.
  • Zan Lynx - Thursday, September 17, 2020 - link

    Drive damage during shipping or infant mortality are both problems you can't avoid.

    I received a couple of bad 15K Seagate SCSI disks one time. This was quite a while ago, they were 36 GB models.

    Stuff happens. It doesn't mean all of those drive models are defective.

Log in

Don't have an account? Sign up now