- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Is there something like the iostat(1) "util%" metric, but that is effective for gauging the amount of load an Optane can take?
In my testing, "util%" is not a good indicator of the maximum load, as it can show 100% utilization with a single-threaded workload, but the Optane can still serve multiple threads simultaneously, which would make the throughput increase meanwhile the util% would stay at 100%. In other words 100% utilization is not a hard limit for the Optane.
That said I have not found another source of information that can suggest if the Optane has or hasn't more bandwidth to offer. I can monitor the read/write throughput or the IOPS but it is not trivial to estimate what the maximum limit would be for any kind of mixed random/sequential workload with operations of differing block sizes.
So, is there some other method of estimating if the performance of an Optane device has peaked or it can increase further?
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi simonebaracchi,
Good day.
Technically we do have an tools that is available to monitor the performance and health of the Optane but this is only applicable to Optane DC SSDs. Refer the link below:
Another way to monitor on the performance is to use the Intel® Optane™ Memory and Storage Management Application. Refer the link below:
Thanks,
Faez_Intel
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello simonebaracchi,
Regarding your thread, we are following up to find out if you were able to complete the actions, we previously recommended.
Please reply here to confirm, so we can continue helping with a resolution. Looking forward to receiving your reply!
Thank You.
Regards,
Hayat
Intel® Customer Support
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello simonebaracchi,
Regarding your thread, we are following up to find out if you were able to complete the actions, we previously recommended.
Please reply here to confirm, so we can continue helping with a resolution. Looking forward to receiving your reply!
Thank You.
Regards,
Hayat
Intel® Customer Support
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Faez, Hayat,
Thank you but neither of these documents seem to address the topic of performance monitoring, and measuring the device utilization in terms of amounts of performance.
I will try some of the troubleshooting advice, but I do not think my devices have performance issues.
My goal would be to understand if the Optane devices are a bottleneck or not in my systems. They show near-100% utilization, but it is unclear whether this is a limit or not.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello simonebaracchi,
Thank you for the update.
Kindly share with us the screenshot where it shows near-100% utilization.
Can you please also share with us the SSU Logs to check further using below link:
https://www.intel.com/content/www/us/en/support/articles/000008563/ethernet-products.html
Regards,
Hayat
Intel Customer Support
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello,
See for example this chart:
This chart represents utilization (as seen via util% in iostat(1)), in purple, and a performance metric, in green, measured during various benchmarks. (For simplicity we can assume the performance metric is proportional to the amount of IOPS)
As you can see, most of the time the purple and green line mostly overlap, which means, if util% is at 100% then performances cannot increase further. But in some instances, highlighted by the red circle, they do not overlap, which means that even if util% is at 100%, we can still get more throughput from the device.
The difference between them is that these within the red circle are single-threaded benchmarks, while those outside of the circles are multi-threaded.
My point is that "util%", which for HDDs is generally a good indicator of the device inability to provide more performance, is not useful for Optanes. From the chart we can see that a single-thread benchmark can push enough requests to be constantly waiting for the device response (pushing the util to 100%), but a second I/O thread can increase the throughput even further. Maybe also a third and fourth and so on.
This is not an issue with the performance, but rather an issue with the monitoring. This is probably common behavior for all SSDs that can serve multiple threads simultaneously (I think my specific model can handle between 2 and 4 threads with almost linear performance increase).
To summarize: given a system with a certain amount of workload, how do I find out how much more performance I can get? If starting more tasks will result in more throughput, or if I have reached the Optane performance limit?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello simonebaracchi,
Thank you for the update. We are currently checking further on this.
May I know below information:
- System name/brand and model
- Operating System
- Optane model
Please be informed for below notes:
The products Intel® Optane™ Memory H10 and H20 with Solid State Storage are now “EOIS” products starting from January 17, 2024. That means, technical support is no longer available for customers from all channels: chat, phone, community, email, web. Warranty support remains unchanged, if the product is still within the 5-year warranty period we provide warranty support.
Regards,
Hayat
Intel Customer Support
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I mostly work with systems that are RHEL9 based but the distro is customized.
I do not have a specific list of hardware devices. This behavior was observed in a system with Optane disks but it may apply to all Optanes and even to all SSDs. This specific test was run on a SSDPE21D480GA.
My original question is, more generically, how to monitor the load on generic Optane disks.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello simonebaracchi,
Thank you for your update.
We are currently checking further on the issue and will get back soon once we know more updates.
Regards.
Hayat
Intel Customer Support
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello simonebaracchi,
Thank you for your patience,
Upon checking, there is no such tool that monitors the performance of Optane SSDs.
Optane drives are NVMe SSDs that behave like any other NVMe SSD. Intel doesn't have any particular way to monitor the drive's performance according to the way you had mentioned.
You are highly suggested to refer and follow the following evaluation guide, but this is more about running a benchmark rather than monitoring performance: Evaluation Guide for Client Intel® Optane™ SSD's
Kindly let us know if you have any other questions.
Regards,
Hayat
Intel Customer Support
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello simonebaracchi,
Regarding your thread, we are following up to find out if you were able to complete the actions, we previously recommended.
Please reply here to confirm, so we can continue helping with a resolution. Looking forward to receiving your reply!
Thank You.
Regards,
Hayat
Intel® Customer Support
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello simonebarrachi,
Since there are no replies, we will proceed to close the thread.
Please be informed, once the thread is closed it will no longer be monitored by Intel.
Kindly open a new thread if you have further questions.
Regards,
Hayat
Intel® Customer Support
![](/skins/images/D3C0B914909A6564BBB97F4AD1ED1973/responsive_peak/images/icon_anonymous_message.png)
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page