Hamid Ali
asked on
How to get 240000 IOPS
Hi friends,
I have a nvme ssd- PM953 960G gen3 sumsung,
it claim has 240000 IOPs random read, and 19000 IOPS random write,
sequential read/write=1GB/780MB
I want test it's IOPS , how i test it for get to IOPS that it said?
what is tools for test and what is configuration for it?
very thanks,
I have a nvme ssd- PM953 960G gen3 sumsung,
it claim has 240000 IOPs random read, and 19000 IOPS random write,
sequential read/write=1GB/780MB
I want test it's IOPS , how i test it for get to IOPS that it said?
what is tools for test and what is configuration for it?
very thanks,
Samsung NVMe 960 SSD drives are very good. That is what I have in my laptop computer.
You can be pretty sure it performs as advertised.
You can run the Samsung smart tests (I use the tests provided by Lenovo) to test the health of the drive.
You can get Samsung tests here under Downloads.
http://www.samsung.com/semiconductor/minisite/ssd/product/consumer/960pro/
You can be pretty sure it performs as advertised.
You can run the Samsung smart tests (I use the tests provided by Lenovo) to test the health of the drive.
You can get Samsung tests here under Downloads.
http://www.samsung.com/semiconductor/minisite/ssd/product/consumer/960pro/
it claim has 240000 IOPs random read, and 19000 IOPS random write,
I'm sure it does, ... but only on the specific testing equipment used by whoever did that benchmark. On a naked motherboard with networking disabled, running a stripped version of linux, reading the data in the most efficient sequence, where nothing is done with the data being read, it's probably possible to get 240K IO/sec.
But it would be remarkable to see that kind of I/O on a Windows system and imo it couldn't be done on a production system running a real application under load.
Benchmarks are only useful when the specific circumstances of how they were done are published. Otherwise it's just pulling data out of thin air.
I'm sure it does, ... but only on the specific testing equipment used by whoever did that benchmark. On a naked motherboard with networking disabled, running a stripped version of linux, reading the data in the most efficient sequence, where nothing is done with the data being read, it's probably possible to get 240K IO/sec.
But it would be remarkable to see that kind of I/O on a Windows system and imo it couldn't be done on a production system running a real application under load.
Benchmarks are only useful when the specific circumstances of how they were done are published. Otherwise it's just pulling data out of thin air.
This question needs an answer!
Become an EE member today
7 DAY FREE TRIALMembers can start a 7-Day Free trial then enjoy unlimited access to the platform.
View membership options
or
Learn why we charge membership fees
We get it - no one likes a content blocker. Take one extra minute and find out why we block content.
ps find the following link ... download the appropriate one..
https://www.raymond.cc/blog/measure-actual-hard-disk-perfomance-under-windows/
all the best