Performance - Internal Storage Mode

The performance of an internal storage device is dependent on the performance characteristics of the device as well as the file system being used. In order to isolate the effects of the latter, we first benchmarked the raw drives using HD Tune Pro 5.70. It was then formatted in NTFS and subject to our standard direct-attached benchmark suite.

Raw Drive Performance

HD Tune Pro allows us to run a variety of tests to determine transfer rates and IOPS for various artificial workloads. In addition, it also allows us to visualize how the performance varies as the tracking head moves from the outer parts of the platter towards the center (i.e, transfer rates as a function of the block address).

Empty drives are bound to perform very well, but, depending on the location of the data in the drive, we find that access rates can go as low as 117 MBps for sequential workloads (compared to the 112 MBps of the 12TB drive from last year). Write access times are a bit unpredictable due to the multi-segmented cache.

Random accesses are never the strong points of hard drives, and we see that the BarraCuda Pro delivers around 69 IOPS for 4K random reads and 222 IOPS for 4K random writes. The random read performance is very similar to the performance of the 12TB version from last year, while the random write performance receives a boost up from 191 IOPS.

The extra tests help in putting some numbers to sequential accesses targeting different areas of the drive. It also provides some interesting numbers relevant to various random access workloads.

DAS Benchmarks

Consumers opting for drives such as the 12TB Seagate BarraCuda Pro typically need high-capacity local storage for holding and editing / processing large-sized multimedia files. Prior to taking a look at the real-life benchmarks, we first check what CrystalDiskMark has to report for the drive. The numbers are in the same ballpark as the 12TB version from last year.

In order to tackle the real-life use-case of transferring large amounts of data back and forth from the drive, we created three test folders with the following characteristics:

  • Photos: 15.6 GB collection of 4320 photos (RAW as well as JPEGs) in 61 sub-folders
  • Videos: 16.1 GB collection of 244 videos (MP4 as well as MOVs) in 6 sub-folders
  • BR: 10.7 GB Blu-ray folder structure of the IDT Benchmark Blu-ray (the same that we use in our robocopy tests for NAS systems)
Seagate BarraCuda Pro 14TB robocopy Benchmarks (MBps)
  Write Bandwidth Read Bandwidth
Photos 197.81 206.81
Videos 219.51 206.99
Blu-ray Folder 222.00 212.36

These numbers are better than the ones we obtained for the 10TB version from 2016, but, fall a little short of the performance of the 12TB drive from last year.

While processing our DAS suite, we also recorded the instantaneous transfer rates and temperature of the drive. Compared to typical disk drives, the write transfers show higher instantaneous speeds due to a combination of the firmware and the 256 MB cache inside the drive. However, sustained write rates are comparable to other high-capacity drives when the cache is exhausted. The temperature of the unit at the end of the transfers (more than 250GB of traffic) rose by less than 4C, pointing to the power-efficiency of the platform.

For the use-case involving editing of multimedia files directly off the disk, we take advantage of PCMark 8's storage benchmark. The storage workload is a good example of a user workload, involving games as well as multimedia editing applications. The command line version allows us to cherry-pick storage traces to run on a target drive. We chose the following traces.

  • Adobe Photoshop (Light)
  • Adobe Photoshop (Heavy)
  • Adobe After Effects
  • Adobe Illustrator

Usually, PCMark 8 reports time to complete the trace, but the detailed log report has the read and write bandwidth figures which we present in our performance graphs. Note that the bandwidth number reported in the results don't involve idle time compression. Results might appear low, but that is part of the workload characteristic.

Seagate BarraCuda Pro 14TB PCMark8 Storage Benchmarks (MBps)
  Write Bandwidth Read Bandwidth
Adobe Photoshop (Light) 248.16 9.01
Adobe Photoshop (Heavy) 233.95 11.03
Adobe After Effects 89.93 8.73
Adobe Illustrator 202.41 8.47

Compared to the results from the 12TB drive last year, we find that the read workloads fare very similarly, but, the write workloads have a slight edge.

Introduction Performance - Direct Attached Storage Mode
POST A COMMENT

70 Comments

View All Comments

  • timecop1818 - Monday, September 10, 2018 - link

    Seagate? Nope. Drives over 4-5TB not in RAID? Nope. I'd like to actually keep my data. Reply
  • Hixbot - Monday, September 10, 2018 - link

    What's preventing you from using RAID with this drive? I have 5 of the 10tb versions in raid 5. Reply
  • wumpus - Monday, September 10, 2018 - link

    Mostly cost, although once you get into 40TB raid land I suspect the convenience of >10TB drives outweighs the issue of buying multiple 8 port SATA pci-e cards to go along with 16 3TB drives. The 3TB still wins in cost (although any more and I'd look into jumping to 4TB, the cost of the ports outweighs the cost of the parity drives) and you can also bump up to RAID 6. On second thought, I'm not sure how I'm going to wrap a case around 16 drives: that might be the biggest hurdle.

    Not that I'm convinced that RAID6 is necessary (as far as I can tell most "you need RAID6" assumed that bad bits were randomly dispersed and that a single bad bit would kill your data, actually bad bits have to be in 4k(byte) chunks (32,000 times as rare as they thought) and that at least consumers aren't going to store anything that can't afford a single [sector] of error (at least at 40TB level). But If I'm already buying 15 hard drives for RAID5, I'll bump it up to RAID6.

    Do you back that monster up on LTO? Picking up refurbished/semi-obsolete LTO for consumer gear appears complicated (but eventually justifies the cost): picking up "list price pro gear" makes more sense to have a backup RAID array.
    Reply
  • close - Monday, September 10, 2018 - link

    It's not related to cost. Someone just knows RAID used to be a buzzword and he has to slip into the conversation to add weight to some dubious opinions.

    Seagate stands a lot better now reliability wise than it used to. Backblaze data seems to back this up. Drives don't "come" in RAID. You build it. Pretty sure he also goes for RAID 5 because it's "better", ""best performance/safety ratio".
    Reply
  • bcronce - Monday, September 10, 2018 - link

    RAID 5 and 6 don't protect from data corruption, just dataloss. ZFS has an custom implementation of RAID 5 and 6 that can detect corruption and figure out which source has an uncorrupted version.

    There are many other reasons than passive data corruption that can mess up your RAID 5/6.
    Reply
  • wumpus - Tuesday, September 11, 2018 - link

    Somehow I don't think I'm building a >40TB array with anything but ZFS (plus its own RAID).

    RAID is overblown and I'd expect more failures from things other than drives self-bricking. RAID's best feature is probably for yanking a drive due to SMART failures or aging out of an array. You'd have to take down a JBOD array to image a new drive from the one aged out. I wouldn't expect you to want to use the data from a drive yanked due to SMART failures. If you are just going to "build and forget" the array, RAID isn't buying you much.

    If your network is up to it (it probably isn't), a good RAID should be able to handle the bandwidth of a large chunk of files being copied from an SSD (and easily should handle it if they are on the same computer, but that seriously limits your filesystem choices. Don't do that).
    Reply
  • MrSpadge - Monday, September 10, 2018 - link

    The value of data on a drive does not strictly depend on its size. Backup if it's important, don't if it's just temp storage or can easily be regenerated. Reply
  • Beaver M. - Tuesday, September 11, 2018 - link

    Im running 2 12 TB not in RAID. Just using one of them as backup. Much safer than a RAID.
    No idea what youre talking about.
    Reply
  • mapesdhs - Wednesday, September 12, 2018 - link

    Are both drives in the same system? Hope you don't get a lightning strike. :}

    Just curious though, what's "safer" about doing it manually than simply using RAID1?
    Reply
  • close - Thursday, September 13, 2018 - link

    RAID 1 is the perfect way to ensure that any fcukup with the data on one drive is promptly replicated on the second one. In most normal cases performing a sync later is a much better option. A little more resistant to errors. Reply

Log in

Don't have an account? Sign up now