The Seagate BarraCuda (500GB) SSD Review: Getting Back In The Game
by Billy Tallis on December 7, 2018 8:00 AM ESTRandom Read Performance
Our first test of random read performance uses very short bursts of operations issued one at a time with no queuing. The drives are given enough idle time between bursts to yield an overall duty cycle of 20%, so thermal throttling is impossible. Each burst consists of a total of 32MB of 4kB random reads, from a 16GB span of the disk. The total data read is 1GB.
The burst random read speed of the Seagate BarraCuda is rather slow, both in comparison to the broad field of competing SSDs, and in comparison to the other drives that use Toshiba's 3D TLC: the Plextor M8V is 30% faster, and even the DRAMless Toshiba TR200 is 15% faster.
Our sustained random read performance is similar to the random read test from our 2015 test suite: queue depths from 1 to 32 are tested, and the average performance and power efficiency across QD1, QD2 and QD4 are reported as the primary scores. Each queue depth is tested for one minute or 32GB of data transferred, whichever is shorter. After each queue depth is tested, the drive is given up to one minute to cool off so that the higher queue depths are unlikely to be affected by accumulated heat build-up. The individual read operations are again 4kB, and cover a 64GB span of the drive.
On the longer random read test, the BarraCuda appropriately pulls ahead of the Toshiba TR200 but remains much slower than the other current-generation mainstream competitors.
 |
|||||||||
| Power Efficiency in MB/s/W | Average Power in W | ||||||||
The BarraCuda is among the most power-hungry drives on this test, leaving it with a very poor power efficiency rating. The best TLC SATA drives offer 50% better performance per Watt, and the Plextor M8V gets 16% better performance while drawing the same amount of power.
 |
|||||||||
As queue depths increase, the BarraCuda stays a little bit slower than most mainstream TLC SSDs, and past QD16 it starts to level out with significantly slower random read performance than the competition. On the bright side, power consumption doesn't increase very quickly as queue depth grows, so the efficiency at high queue depths is no longer such a negative outlier.
Comparing the Seagate BarraCuda against our entire collection of SATA SSDs new and old, it's clear that the power efficiency could use some improvement but it's by no means the worst we've tested.
Random Write Performance
Our test of random write burst performance is structured similarly to the random read burst test, but each burst is only 4MB and the total test length is 128MB. The 4kB random write operations are distributed over a 16GB span of the drive, and the operations are issued one at a time with no queuing.
The Seagate BarraCuda disappoints on the burst random write test with QD1 performance that is well below that of many current and previous generation SSDs, including some DRAMless drives. The Plextor M8V is 36% faster on this test due to pairing the same Toshiba NAND with a more recent Silicon Motion controller rather than the Phison S10.
As with the sustained random read test, our sustained 4kB random write test runs for up to one minute or 32GB per queue depth, covering a 64GB span of the drive and giving the drive up to 1 minute of idle time between queue depths to allow for write caches to be flushed and for the drive to cool down.
On the longer random write test that involves some higher queue depths, the differences between mainstream SATA drives are smaller than at QD1 but the BarraCuda is still clearly below par. However, it is at least far faster than the DRAMless drives.
 |
|||||||||
| Power Efficiency in MB/s/W | Average Power in W | ||||||||
Power consumption for the BarraCuda on this test is a bit high, so the efficiency score is again fairly low. Other current mainstream SATA SSDs offer at least 20% better performance per Watt. The Plextor M8V is 18% faster while drawing 12% less power, and that still isn't close to the best efficiency we're seeing from non-Toshiba 3D TLC.
 |
|||||||||
Random write performance saturating at QD4 and higher is normal for SATA drives, but the performance level that the BarraCuda tops out at is well below what we expect from mainstream SSDs these days. The Plextor M8V delivers more performance at lower power levels across all queue depths, while using a 4-channel Silicon Motion controller instead of the 8-channel Phison S10 to manage the same Toshiba 3D TLC.
Comparing the Seagate BarraCuda against every SATA SSD that has completed this test, we again see that power consumption is substantially higher than the best drives out there, and the maximum random write performance attained falls far short of the best that a SATA drive can deliver.
Facebook
Twitter
RSS
39 Comments
View All Comments
seamonkey79 - Friday, December 7, 2018 - link
The caddy would adapt an m.2 to a full 2.5" SATA socket, so you would have an m.2 in a chassis adapting it to 2.5" & classic SATA.That being said, I can't see that doing anything but increasing costs, though having one primary line to manufacture m.2 and the little bit of work needed to adapt an m.2 SATA drive to 2.5", it *could* see some benefit to the manufacturer.
Death666Angel - Friday, December 7, 2018 - link
Huh? M.2 SATA doesn't overheat, that's M.2 NVME/PCIe (which is a bit more difficult to adapt to 2.5" SATA, so I don't care as much). 2.5" SATA drives already have tiny PCBs. They are already bottlenecked by the protocol. With them being M.2 in a 2.5" caddy, they can serve double purpose. I just bought an M.2 drive and a caddy for my Fujitsu T904 laptop, which still has only a 2.5" slot. But I know when that laptop is gone, I'm not gonna need a 2.5" drive. That M.2 drive can be converted into all kinds of useful devices, small desktops, laptops, USB thumb drive. It'd cost them a couple dozens of cents more to manufacture, would it'd be soo much more useful. :D Eventually, more people will use M.2 SATA than 2.5". At least those who buy standalone drives.CheapSushi - Saturday, December 8, 2018 - link
Dragonstongue I don't think you really know what the OP is talking about, just how tiny the PCB seems to be on newer 2.5" SSDs, thus just being a whole lot of waste of space anyway and/or you don't realize adapters already exist and/or don't realize M.2 is just a formfactor and can be NVMe/PCIe or AHCI/SATA. There's always someone who confidently posts a "neg" at a suggestion without even knowing hardware much in the first place.dgingeri - Friday, December 7, 2018 - link
When I was a systems admin for a server software test lab, we received a set (108) of 3TB Seagate Constellation ES.2 SAS drives for a new prototype appliance that would eventually become the DXi6900 series. I was pretty excited to see the new hardware come in, so I got things set up within a day.The test team wasn't ready to test for another week. In that week, we had 12 of those drives go bad. By two months into the project, every single Seagate drive had gone bad. (These are their enterprise level drives, which are supposed to have better reliability.) We ended up requesting a different brand drive from NetApp (the maker of the storage portion of the DXi6900) because of these problems. We replaced them with HGST 4TB drives, and didn't have a single one fail up to the point when I left over two years later.
In the years leading up to that, I had bought several Seagate drives, including 4 1TB drives, 2 2TB drives, and one 3TB drive, and had the drive fail within warranty in EVERY SINGLE CASE. That was specifically why I quit buying them, and the 3TB drive is the last Seagate drive I am ever going to buy intentionally.
This just might be a decent crive, and if Seagate were to put a concerted effort into improving their reliability, they might be something I'd consider. However, as things stand, Seagate and Toshiba are on my NEVER BUY list, along side Biostar, ECS, and Gigabyte.
CheapSushi - Saturday, December 8, 2018 - link
What does this have to do with SSDs at all? Seriously? This is just some overreaching rant saying that one specific type of product means that ALL their products are a problem. Especially about spinning rust several years old; even BackBlaze doesn't use those. Have you even bothered to look into reliability statistics/information on even consumer TLC drives? It's so odd to see people who claim to be in tech but are so antiquated in their knowledge.gglaw - Sunday, December 9, 2018 - link
Quite a humorous post from someone with a supposed heavy tech/admin background but so short-sighted on the big picture. He lists a bad experience from a completely unrelated product line likely not even sharing manufacturing or R&D ties in any way making him ban products from some of the largest tech companies in the world with for the most part tremendous track records. Even if it is in "principle" for how the company leaders model their QC, all the executives making these decisions at the time of the archaic hard drive problems are likely working with other companies by now (pretty good chance for one of his "new" favorite companies). Similar to the comments on some of Samsung's early SSD fiascos banning all Samsung products "for life." And of course shortly after their fiascos, they quickly became essentially the world's benchmark for performance and reliability in this same product line lol.Donkey2008 - Thursday, December 13, 2018 - link
108 enterprise-class hard drives failed in 2 months?[Insert Doubt meme]
sarahkevin - Friday, December 7, 2018 - link
thanks for sharing I really need this for my office.PeachNCream - Friday, December 7, 2018 - link
I think I'd probably opt for a Crucial MX500 over a Seagate SSD. Seagate's reputation and my experiences with their mechanical drives make me reluctant about giving them yet another chance.Fujikoma - Friday, December 21, 2018 - link
I feel the same way about Quantum SCSI drives. Not that Seagate rates much higher...