Random Read Performance

One of the major changes in our 2015 test suite is the synthetic Iometer tests we run. In the past we used to test just one or two queue depths, but real world workloads always contain a mix of different queue depths as shown by our Storage Bench traces. To get the full scope in performance, I'm now testing various queue depths starting from one and going all the way to up to 32. I'm not testing every single queue depth, but merely how the throughput scales with the queue depth. I'm using exponential scaling, meaning that the tested queue depths increase in powers of two (i.e. 1, 2, 4, 8...). 

Read tests are conducted on a full drive because that is the only way to ensure that the results are valid (testing with an empty drive can substantially inflate the results and in reality the data you are reading is always valid rather than full of zeros). Each queue depth is tested for three minutes and there is no idle time between the tests. 

I'm also reporting two metrics now. For the bar graph, I've taken the average of QD1, QD2 and QD4 data rates, which are the most relevant queue depths for client workloads. This allows for easy and quick comparison between drives. In addition to the bar graph, I'm including a line graph, which shows the performance scaling across all queue depths. To keep the line graphs readable, each drive has its own graph, which can be selected from the drop-down menu.

I'm also plotting power for SATA drives and will be doing the same for PCIe drives as soon as I have the system set up properly. Our datalogging multimeter logs power consumption every second, so I report the average for every queue depth to see how the power scales with the queue depth and performance.

Iometer - 4KB Random Read

Random read performance at small queue depths has never been an area where the Vector 180 has excelled in. Given that these are one of the most common IOs, it's an area where I would like to see improvement on OCZ's behalf.

Iometer - 4KB Random Read (Power)

Power consumption, on the other hand, is excellent, which is partially explained by the lower performance. 

Samsung SM951 512GB

Having a closer look at the performance data across all queue depths reveals the reason for Vector 180's poor random read performance. For some reason, the performance only starts to scale properly after queue depth of 4, but even then the scaling isn't as aggressive as on some other drives. 

Random Write Performance

Write performance is tested in the same way as read performance, except that the drive is in a secure erased state and the LBA span is limited to 16GB. We already test performance consistency separately, so a secure erased drive and limited LBA span ensures that the results here represent peak performance rather than sustained performance.

Iometer - 4KB Random Write

In random write performance the Vector 180 does considerably better, although it's still not the fastest drive around. 

Iometer - 4KB Random Write (Power)

Even though the random write performance doesn't scale at all with capacity, the power consumption does. Still, the Vector 180 is quite power efficient compared to other drives.

Samsung SM951 512GB

The Vector 180 scales smoothly across all queue depths, but it could scale a bit more aggressively because especially the QD4 score is a bit low. On a positive side, the Vector 180 does very well at QD1, though.

AnandTech Storage Bench - Light Sequential Performance
POST A COMMENT

89 Comments

View All Comments

  • Shark321 - Wednesday, March 25, 2015 - link

    Tosh, it's a pity PFM does not work on the internal cache of the drive. You can still get file system damage during a power loss. Reply
  • AVN6293 - Sunday, December 20, 2015 - link

    Does this drive support Opal 2.0 eDrive (FIPS/Hippa compliance) ? Reply
  • AVN6293 - Sunday, December 20, 2015 - link

    ...And can the over provisioning be increased by the user ? Reply
  • ats - Wednesday, March 25, 2015 - link

    Actually, all consumer drives need power loss protection and they realistically need it much more than drives targeted at the actual enterprise side of the market. It comes down to simple probabilities. The average enterprise SSD is going to be backed by at least 1 additional layer of power loss prevention (UPS et al), have a robust backup infrastructure, and likely mirroring (offsite) on top.

    In contrast, consumer drives are unlikely to have any power loss prevention, unlikely to have anything approaching a backup infrastructure, and highly unlikely to have robust data resiliency(offsite mirroring et al).

    So like many others, Anandtech gets it exactly wrong wrt PLP and SSDs. The fact that manufacturers have been able to get away without providing PLP on consumer SSDs is almost criminal. The fact that review sites accept this as perfectly OK is pretty much criminal on their part.

    And what should pretty much be a rage storm for consumers is the actual cost of providing PLP on an SSD is literally a couple of $ in capacitors. Not to mention many consumer drives without PLP have enterprise drives using the exact same PCB with PLP. That we as consumers have allowed companies to have PLP as a point of differentiation is to our great detriment, esp when the actual cost of PLP is in the noise even for cheap low capacity SSDs.

    If a drive cannot survive a power loss with data integrity then it certainly shouldn't get a recommendation nor should any consumer even consider it.
    Reply
  • Shiitaki - Wednesday, March 25, 2015 - link

    You do raise some very good points. I think the enterprise still needs it because they want as many ways to protect the data that they can get, after all it's only a couple of bucks. The consumer would benefit to a greater degree since that is likely all they would have is the caps in the SSD. However the consumer is their own worse enemy, a couple of bucks makes a difference for most consumers.

    I've had no issues, and until I read this article, gave no thought to pulling power on a system using an SSD! And I've done it ALOT! Not a single bad block yet! And that is with 6 SSD's in various machines from 4 manufacturers and 8 product lines. Though none of them with Windows, all Linux and OS X.

    Sometimes I wonder just how wide spread issues really are. On the internet it's hard to tell since it's the angry people doing most of the posting.

    In the end though whether you area company or individual, if it isn't backed up. you really don't need it.
    Reply
  • trparky - Wednesday, April 29, 2015 - link

    I do an image of my system SSD every week and my computer is always plugged into a UPS, and yes, that's my home setup. The power is my area is known to be dirty power, not complete drop-outs but if you measured the voltage output it would make most electrical engineers shake their heads and smack their foreheads. Reply
  • zodiacsoulmate - Tuesday, March 24, 2015 - link

    also the Arconis 2013 is basically useless since it only runs on windows 7.... Reply
  • ocztosh - Tuesday, March 24, 2015 - link

    Hi Zodiassoulmate, just wanted to confirm that the Vector 180 drives are shipping with Acronis 2014. Reply
  • DanNeely - Tuesday, March 24, 2015 - link

    That's a step in the right direction; but is still last years product. Acronis 2015 is already out. Am I overly cynical for thinking Acronis offered the 2014 version at a discount hoping to make it up by convincing some of the SSD buyers to upgrade to the new version after installing? Reply
  • Samus - Tuesday, March 24, 2015 - link

    Acronis TrueImage 2015 is complete shit. Check the Acronis forums: most people (like myself - a paying annual customer since 2010) have gone back to 2014. The most recent update (October) still did not fix issues with image compatibility, GPT partition compatibility (added for 2015) and UEFI boot mapping. Aside from the lingering compatibility, reliability and stability issues, the interface is terrible. They've basically turned it into a backup product for single PC's instead of a imaging product. Even the USB bootable ISO I typically boot off a flash drive for imaging/cloning is inherently unstable and occasionally even corrupts the destination. Nobody has confirmed the "Universal Restore" works for Windows 7, yet another broken feature that worked FINE in 2014.

    Acronis lost me as a long-time customer to Miray because 2015 was SO botched and after waiting months for them to fix it, I gave up and had to find a product that could adequately clone UEFI OS's installed on GPT partitions. I use this product almost daily to upgrade PC's to SSD's. Unfortunately Miray's boot environment is a little slower, even with the verification disabled and "fast copy" turned on, likely because it runs a different USB stack.

    I don't blame OCZ for sticking with 2014 like every other Acronis licensee has, including Crucial and Intel. 2014 is mature and stable, but it is not the modern solution - especially with Windows 10 around the corner. Acronis will forfeit this market to Miray or in-house solutions like Samsungs' Clonemaster if they don't get their act together. It's just astonishing how well Acronis was doing until 2015.
    Reply

Log in

Don't have an account? Sign up now