ASRock Forums : RAID 0 NVME Read Performance Same as Non-RAID

RAID 0 NVME Read Performance Same as Non-RAID : Thank you for explaining. I feel...

Wed, 21 May 2025 03:03:32 +0000

Author: Caxton
Subject: 107182
Posted: 21 May 2025 at 3:03am

Thank you for explaining. I feel less neurotic (than normal). The additional context is helpful.

It appears RAID 0 read scaling remains an issue.

ASRock has separately confirmed that RAID 0 read performance on this system board is seemingly impaired, failing to function as expected. Even with two identical Gen 4 NVMe SSDs in a RAID 0 array, read operations exhibit performance equivalent to that of a single drive.

I have yet to find any discussions reporting successful RAID 0 read and write scaling on this platform. Specifically, two Gen 4 NVMe SSDs performing as expected in a typical RAID 0 setup. This includes comparisons with other chipset-based RAID implementations that do not require add-in RAID controllers. If this limitation is inherent to the board, it suggests misleading marketing of its RAID 0 NVMe support. The absence of read scaling effectively negates the core advantage of RAID 0, where performance should scale proportionally with the number of drives (e.g., at two times for two SSDs, three times for three, etc.).

RAID 0 NVME Read Performance Same as Non-RAID : Caxton wrote:Thank you all...

Wed, 30 Apr 2025 14:25:27 +0000

Author: M440
Subject: 107182
Posted: 30 Apr 2025 at 2:25pm

Caxton wrote:

Thank you all for helping debug this issue.

It appears the initial suggestion was AI-generated, based on the style and format in the original post that referenced "Can I use a Fedora Live CD to mount...". While it was helpful to some extent, the proposed solution ultimately did not work as expected due to several limitations. Fedora LiveCD/USB environments come in various versions, and commands can behave differently depending on the specific combination of version and environment used.

For example, the following command failed when executed in the Fedora 42 Live environment (booted via USB media):

sudo dnf install dmraid

The failure message was:
"failed to resolve the transaction" and "no match for argument dmraind:"

To provide additional context: Fedora 42's Live CD/USB environment does not include `dmraid` by default, nor does it allow installation via the above command. A subsequent search within the booted environment, executed as follows:

sudo dnf search dmraid
...returned:
"no matches found."

The search was performed after the install command failed, simply to confirm whether the `dmraid` package might already exist in the environment?�but it does not.

So, since this approach appears to be a dead end (at least thus far), are there any alternative methods to non-destructively test RAID 0 read performance?

For reference, when testing in the native environment (Windows 11), the issues are consistently related to read commands, irrespective of hardware connection configurations. However, write commands only produce at RAID 0 performance thresholds when one SSD is connected directly to the CPU bus and the other is on the chipset bus. These findings have been consistent across all tested configurations.

im really am sorry, i literally asked the chatbot that in next prompts and it assured me that fedora should be the most straight forward to do that and wont need internet connection. Personally i just use arch

below are the results i get on a single kc3000,nv2 drives:


> sudo hdparm -Tt /dev/nvme0n1p4

/dev/nvme0n1p4:
 Timing cached reads:   57590 MB in  1.99 seconds = 28908.02 MB/sec
 Timing buffered disk reads: 7560 MB in  3.00 seconds = 2519.50 MB/sec
mm@desktop ~
> sudo hdparm -Tt /dev/nvme1n1p5

/dev/nvme1n1p5:
 Timing cached reads:   58770 MB in  1.99 seconds = 29503.06 MB/sec
 Timing buffered disk reads: 8126 MB in  3.00 seconds = 2708.63 MB/sec

Edited by M440 - 30 Apr 2025 at 2:25pm

RAID 0 NVME Read Performance Same as Non-RAID : Regarding RAID 0 performance,...

Wed, 30 Apr 2025 10:34:18 +0000

Author: NDRE28
Subject: 107182
Posted: 30 Apr 2025 at 10:34am

Regarding RAID 0 performance, are you achieving close to 2x read/write speeds compared to a single SSD? This presumes at least one array is in a 2x NVMe/SSD RAID 0 config.

Hi!
I am achieving more than double the speed of 2 NVMe drives.
However, please keep in mind that I am using 3 identical drives: Samsung 970 Pro 1TB SSDs (PCIe 3.0 x4).
Obtaining double the speed of PCIe 4.0 drives, through RAID-0, is harder.

My primary goal is to build a RAID 0 array using two Samsung 990 Pro 4TB NVMe SSDs, resulting in a single 8TB volume optimized for read/write performance in line with the hardware capabilities?�specifically, nearly doubling bandwidth for reads and writes.

In real world usage, one Samsung 9100 Pro 8TB SSD (PCIe 5.0 x4) will be faster than 2x Samsung 990 Pro drives in RAID-0.
That's because RAID setups add some latency, though the sequential transfer speeds go up.
[Please keep in mind that the 8TB version of 9100 Pro hasn't been released yet].

I went to the moon & back with this storage performance thing.
Finally, I came to the conclusion that sequential speeds don't matter so much.
The biggest limitation of these SSDs is their NAND (3-bit per Cell).
Going with a Gen4 Intel Optane, like the DC P5800X, or a Gen4 SLC (1-bit/Cell) drive, like the Solidigm D7-P5810, will lead to better results.

RAID 0 NVME Read Performance Same as Non-RAID : NDRE28 wrote:I am running my...

Wed, 30 Apr 2025 08:09:47 +0000

Author: Caxton
Subject: 107182
Posted: 30 Apr 2025 at 8:09am

NDRE28 wrote:

I am running my RAID-0 setup on an AMD AM5 platform.
One of my SSDs is connected directly to the CPU, while the other 2 are connected to the chipset.
OS: Windows 11 Enterprise v24H2.

In my case, everything runs as expected, in terms of performance.
RAID-0 did improve the transfer speeds.

Regarding RAID 0 performance, are you achieving close to 2x read/write speeds compared to a single SSD? This presumes at least one array is in a 2x NVMe/SSD RAID 0 config. Ideally, the performance should approach the theoretical maximum achievable across the bus and scale as additional, identical SSD's are added to a RAID 0 array.

I?�m particularly interested in whether you?�re obtaining optimal results, especially in comparison to my own setup. Interestingly, I previously purchased (and later returned) an 8TB WD NVMe drive that exhibited slower performance than a single Samsung NVMe. Theoretically, it should have been significantly slower than two Samsung NVMe drives configured in RAID 0, yet the actual performance difference was negligible.

At present, the results seem inconclusive. I?�d love to hear your insights on this.

My primary goal is to build a RAID 0 array using two Samsung 990 Pro 4TB NVMe SSDs, resulting in a single 8TB volume optimized for read/write performance in line with the hardware capabilities?�specifically, nearly doubling bandwidth for reads and writes. Theoretically, this setup should achieve read speeds of approximately 14,900 MB/s and write speeds of about 13,800 MB/s on par with Gen 5x4 single non-RAID NVMe's

However, RAID 0 read performance has failed to effectively utilize parallelization, with read speeds mimicking single-drive performance as measured by both Windows 11?�s winsat tool and CrystalDiskMark. For instance, across three CrystalDiskMark test runs?�each consisting of five repetitions?�the average read performance was 7,211 MB/s, irrespective of the slot configuration.

Interestingly, write performance closely aligned with real-world RAID 0 expectations, reaching approximately 13,748 MB/s when utilizing the M2_1 and M2_2 slots with PCIe Gen 5x4 with the Gen 4x4 configuration. In contrast, other connection and bus combinations delivered suboptimal write performance, averaging around 6,961 MB/s. This includes all tested physical configurations, such as M2_2 with M2_3 (PCIe Gen 4x4) and M2_3 with M2_4 (PCIe Gen 4x4).

RAID 0 NVME Read Performance Same as Non-RAID : I am running my RAID-0 setup on...

Tue, 29 Apr 2025 10:16:36 +0000

Author: NDRE28
Subject: 107182
Posted: 29 Apr 2025 at 10:16am

I am running my RAID-0 setup on an AMD AM5 platform.
One of my SSDs is connected directly to the CPU, while the other 2 are connected to the chipset.
OS: Windows 11 Enterprise v24H2.

In my case, everything runs as expected, in terms of performance.
RAID-0 did improve the transfer speeds.

RAID 0 NVME Read Performance Same as Non-RAID : Thank you all for helping debug...

Tue, 29 Apr 2025 07:58:28 +0000

Author: Caxton
Subject: 107182
Posted: 29 Apr 2025 at 7:58am

Thank you all for helping debug this issue.

It appears the initial suggestion was AI-generated, based on the style and format in the original post that referenced "Can I use a Fedora Live CD to mount...". While it was helpful to some extent, the proposed solution ultimately did not work as expected due to several limitations. Fedora LiveCD/USB environments come in various versions, and commands can behave differently depending on the specific combination of version and environment used.

For example, the following command failed when executed in the Fedora 42 Live environment (booted via USB media):

sudo dnf install dmraid

The failure message was:
"failed to resolve the transaction" and "no match for argument dmraind:"

To provide additional context: Fedora 42's Live CD/USB environment does not include `dmraid` by default, nor does it allow installation via the above command. A subsequent search within the booted environment, executed as follows:

sudo dnf search dmraid
...returned:
"no matches found."

The search was performed after the install command failed, simply to confirm whether the `dmraid` package might already exist in the environment?�but it does not.

So, since this approach appears to be a dead end (at least thus far), are there any alternative methods to non-destructively test RAID 0 read performance?

For reference, when testing in the native environment (Windows 11), the issues are consistently related to read commands, irrespective of hardware connection configurations. However, write commands only produce at RAID 0 performance thresholds when one SSD is connected directly to the CPU bus and the other is on the chipset bus. These findings have been consistent across all tested configurations.

RAID 0 NVME Read Performance Same as Non-RAID : Good point on the random R/W values....

Tue, 29 Apr 2025 01:38:26 +0000

Author: Caxton
Subject: 107182
Posted: 29 Apr 2025 at 1:38am

Good point on the random R/W values. It seems like it should on this system when compared to another similar setup. Thank you for helping validate the test scenario.

RAID 0 NVME Read Performance Same as Non-RAID : I will give that a go and reply...

Tue, 29 Apr 2025 01:35:57 +0000

Author: Caxton
Subject: 107182
Posted: 29 Apr 2025 at 1:35am

I will give that a go and reply with results. Great suggestion!

RAID 0 NVME Read Performance Same as Non-RAID : Hi! I don't know why this...

Sat, 26 Apr 2025 13:19:27 +0000

Author: NDRE28
Subject: 107182
Posted: 26 Apr 2025 at 1:19pm

Hi!

I don't know why this is so on your system.

I am running 3x Samsung 970 Pro 1TB SSDs in RAID-0, on an ASRock X670E Taichi motherboard, with the same BIOS version & RAID driver as you.

In my case, the sequential speeds are more than double, in benchmarks!

In CrystalDiskMark: Seq.Read=9000+MB/s & Seq.Write=8000+MB/s.
The Random Q1T1 Write=400+MB/s & Random Q1T1 Read=60+MB/s.

Unfortunately, Random Read & Write at Q1T1 is what matters the most!
(That's why I'll go with a DC P5800X Optane drive as my boot drive, as soon as it will arrive).

RAID 0 NVME Read Performance Same as Non-RAID : the nvme drives firmware is updated?...

Sat, 26 Apr 2025 12:01:43 +0000

Author: M440
Subject: 107182
Posted: 26 Apr 2025 at 12:01pm

the nvme drives firmware is updated?

this is BIOS/UEFI RAID?

anyway i would blame:
-os/drivers
-nvme drives firmware
-motherboard/bios/hardware