Difference between revisions of "Hardware"

Jump to navigation Jump to search
1,017 bytes added ,  22:09, 27 June 2020
m
Added note about TRIM on Optane
m (Add section for Optane / 3D XPoint SSDs)
m (Added note about TRIM on Optane)
(3 intermediate revisions by the same user not shown)
Line 177: Line 177:
* Toshiba ZD6300
* Toshiba ZD6300
* Seagate Nytro 5000 M.2 (XP1920LE30002 tested; '''''read notes below before buying''''')
* Seagate Nytro 5000 M.2 (XP1920LE30002 tested; '''''read notes below before buying''''')
** Inexpensive 22110 M.2 enterprise drive using consumer MLC that is optimized for read mostly workloads.
** Inexpensive 22110 M.2 enterprise drive using consumer MLC that is optimized for read mostly workloads. It is not a good choice for a SLOG device, which is a write mostly workload.
** The [https://www.seagate.com/www-content/support-content/enterprise-storage/solid-state-drives/nytro-5000/_shared/docs/nytro-5000-mp2-pm-100810195d.pdf manual] for this drive specifies airflow requirements. If the drive does not receive sufficient airflow from case fans, it will overheat at idle. It's thermal throttling will severely degrade performance such that write throughput performance will be limited to 1/10 of the specification and read latencies will reach several hundred milliseconds. Under continuous load, the device will continue to become hotter until it suffers a "degraded reliability" event where all data on at least one NVMe namespace is lost. The NVMe namespace is then unusable until a secure erase is done. Even with sufficient airflow under normal circumstances, data loss is possible under load following the failure of fans in an enterprise environment. Anyone deploying this into production in an enterprise environment should be mindful of this failure mode.
** The [https://www.seagate.com/www-content/support-content/enterprise-storage/solid-state-drives/nytro-5000/_shared/docs/nytro-5000-mp2-pm-100810195d.pdf manual] for this drive specifies airflow requirements. If the drive does not receive sufficient airflow from case fans, it will overheat at idle. It's thermal throttling will severely degrade performance such that write throughput performance will be limited to 1/10 of the specification and read latencies will reach several hundred milliseconds. Under continuous load, the device will continue to become hotter until it suffers a "degraded reliability" event where all data on at least one NVMe namespace is lost. The NVMe namespace is then unusable until a secure erase is done. Even with sufficient airflow under normal circumstances, data loss is possible under load following the failure of fans in an enterprise environment. Anyone deploying this into production in an enterprise environment should be mindful of this failure mode.
** Those who wish to use this drive in a low airflow situation can workaround this failure mode by placing a passive heatsink such as [https://smile.amazon.com/gp/product/B07BDKN3XV this] on the NAND flash controller. It is the chip under the sticker closest to the capacitors. This was tested by placing the heatsink over the sticker (as removing it was considered undesirable). The heatsink will prevent the drive from overheating to the point of data loss, but it will not fully alleviate the overheating situation under load without active airflow. A scrub will cause it to overheat after a few hundred gigabytes are read. However, the thermal throttling will quickly cool the drive from 76 degrees Celsius to 74 degrees Celsius, restoring performance.
** Those who wish to use this drive in a low airflow situation can workaround this failure mode by placing a passive heatsink such as [https://smile.amazon.com/gp/product/B07BDKN3XV this] on the NAND flash controller. It is the chip under the sticker closest to the capacitors. This was tested by placing the heatsink over the sticker (as removing it was considered undesirable). The heatsink will prevent the drive from overheating to the point of data loss, but it will not fully alleviate the overheating situation under load without active airflow. A scrub will cause it to overheat after a few hundred gigabytes are read. However, the thermal throttling will quickly cool the drive from 76 degrees Celsius to 74 degrees Celsius, restoring performance.
Line 239: Line 239:
= Optane / 3D XPoint SSDs =
= Optane / 3D XPoint SSDs =


These are SSDs with far better latencies and write endurance than NAND flash SSDs. They are byte addressable, such that ashift=9 is fine for use on them. Unlike NAND flash SSDs, they do not require any special power failure protection circuitry for reliability. However, they cost more per GB than NAND flash (as of 2020). They make excellent SLOG devices.
These are SSDs with far better latencies and write endurance than NAND flash SSDs. They are byte addressable, such that ashift=9 is fine for use on them. Unlike NAND flash SSDs, they do not require any special power failure protection circuitry for reliability. There is also no need to run TRIM on them. However, they cost more per GB than NAND flash (as of 2020). The enterprise models make excellent SLOG devices. Here is a list of models that are known to perform well:
 
* Intel DC P4800X[https://www.servethehome.com/intel-optane-hands-on-real-world-benchmark-and-test-results/]
** This gives basically the highest performance you can get as of June 2020.
 
Also, at time of writing in June 2020, only one model is listed. This is due to there being few such drives on the market. The client models are likely to be outperformed by well configured NAND flash drives, so they have not been listed (although they are likely cheaper than NAND flash). More will likely be added in the future.
 
Note that SLOG devices rarely have more than 4GB in use at any given time, so the smaller sized devices are generally the best choice in terms of cost, with larger sizes giving no benefit. Larger sizes could be a good choice for other vdev types, depending on performance needs and cost considerations.


= Power =
= Power =
Editor
348

edits

Navigation menu