Raspberry Pi 5 with M2/PCI SSD drive

[Sergei Grichine]

Hi All,

I upgraded my RPi5 on Dragger, replacing the SD card with an SSD.

Components used:

https://www.amazon.com/dp/B0CHNP7P89

https://www.amazon.com/dp/B0CPPGGDQT

https://www.amazon.com/dp/B0C5D6C1YQ

First I used SD Imager on a Windows 10 machine to backup my SD card.

Then Balena Etcher on the same machine to transfer that image to SSD, placed on the M2/USB adapter.

Then I moved SSD to the RPi Hat, and it booted right away, automatically extending the root partition to full ~256GB.

I probably could've done imaging on the Ubuntu desktop using "dd".

Pros: The RPi feels a bit faster, and I hope for better reliability in the long run.

Cons: I won't be able to save the whole OS image anymore, as imaging a 256GB drive isn't practical.

Note: RPi5 has a "Power" jumper and Ubuntu 24.04 supports safe shutdown when a button is connected to it.

Here is how it looks:

https://photos.app.goo.gl/eAdKiD7YYnL9Vou6A

--

Best Regards,

-- Sergei

[Marco Walther]

On 11/2/24 15:14, Sergei Grichine wrote:
> Hi All,
>
> I upgraded my RPi5 on Dragger, replacing the SD card with an SSD.
>
> Components used:

> https://www.amazon.com/dp/B0CHNP7P89 <https://www.amazon.com/dp/B0CHNP7P89>
> https://www.amazon.com/dp/B0CPPGGDQT <https://www.amazon.com/dp/B0CPPGGDQT>
> https://www.amazon.com/dp/B0C5D6C1YQ <https://www.amazon.com/dp/B0C5D6C1YQ>

I'm using a 'a bit smaller' version with the official cooling fan;-)
https://www.amazon.com/gp/product/B0CQ4D2C9S
https://www.amazon.com/gp/product/B0BG829J8V

Works well so far. But I have cases around the Pi's, so not much to see
any more;-)

One on my rover and one onthe back of a 7" display with case.

Thanks,
-- Marco

>
> First I used SD Imager on a Windows 10 machine to backup my SD card.
> Then Balena Etcher on the same machine to transfer that image to SSD,
> placed on the M2/USB adapter.
> Then I moved SSD to the RPi Hat, and it booted right away, automatically
> extending the root partition to full ~256GB.
>
> I probably could've done imaging on the Ubuntu desktop using "dd".
>
> Pros: The RPi feels a bit faster, and I hope for better reliability in
> the long run.
> Cons: I won't be able to save the whole OS image anymore, as imaging a
> 256GB drive isn't practical.
>
> Note: RPi5 has a "Power" jumper and Ubuntu 24.04 supports safe shutdown
> when a button is connected to it.
>
> Here is how it looks:
>
> https://photos.app.goo.gl/eAdKiD7YYnL9Vou6A
> <https://photos.app.goo.gl/eAdKiD7YYnL9Vou6A>
>
> --
> Best Regards,
> -- Sergei
>

> --
> You received this message because you are subscribed to the Google
> Groups "HomeBrew Robotics Club" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to hbrobotics+...@googlegroups.com
> <mailto:hbrobotics+...@googlegroups.com>.
> To view this discussion visit
> https://groups.google.com/d/msgid/hbrobotics/CA%2BKVXVPg5GLWUmFoz2GUFoZ-0G-3MGbg_0CZcTBpb64qUYOivw%40mail.gmail.com <https://groups.google.com/d/msgid/hbrobotics/CA%2BKVXVPg5GLWUmFoz2GUFoZ-0G-3MGbg_0CZcTBpb64qUYOivw%40mail.gmail.com?utm_medium=email&utm_source=footer>.

[James H Phelan]

S,

Since it's no more memory than an SD card (256GB) must be the speed.

Are you seeing an appreciable difference?

J

James H Phelan
"Nihil est sine ratione cur potius sit quam non sit"
Leibniz

--
You received this message because you are subscribed to the Google Groups "HomeBrew Robotics Club" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hbrobotics+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/hbrobotics/CA%2BKVXVPg5GLWUmFoz2GUFoZ-0G-3MGbg_0CZcTBpb64qUYOivw%40mail.gmail.com.

[Marco Walther]

On 11/3/24 04:24, 'James H Phelan' via HomeBrew Robotics Club wrote:
> S,
>
> Since it's no more memory than an SD card (256GB) must be the speed.

That's just a choice;-) There are 4TB SSD's out there, which would fit
into the 2280 format.

>
> Are you seeing an appreciable difference?

Yes;-)

The [Micro-]SD card controllers are designed to do large sequential
writes & reads (now 4k-video or better). Not for much more than that.

OK, a SanDisk Ultra 32GB (10) A1 U-1 MircoSD against a SSD still in
PCIe-2 mode (apparently the Pi5 can do PCIe-3, they now do it for the AI
Hat by default) shows about a 3x in write performance;-) Normally, a lot
of that is hidden (buffered) from the user, but even 8GB memory is not
that much in some cases;-)

(My laptop with PCIe-4, does about the same for the SD, but about 10x of
the Pi's SSD)

-- Marco

# MicroSD
ubuntu@ubuntu-2204-rpi5:/media/ubuntu/rootfs/tmp$ iozone -t1 -i0 -i2
-r1k -s1g .
Iozone: Performance Test of File I/O
Version $Revision: 3.489 $
Compiled for 64 bit mode.
Build: linux

Contributors:William Norcott, Don Capps, Isom Crawford, Kirby
Collins
Al Slater, Scott Rhine, Mike Wisner, Ken Goss
Steve Landherr, Brad Smith, Mark Kelly, Dr. Alain CYR,
Randy Dunlap, Mark Montague, Dan Million, Gavin
Brebner,
Jean-Marc Zucconi, Jeff Blomberg, Benny Halevy,
Dave Boone,
Erik Habbinga, Kris Strecker, Walter Wong, Joshua
Root,
Fabrice Bacchella, Zhenghua Xue, Qin Li, Darren
Sawyer,
Vangel Bojaxhi, Ben England, Vikentsi Lapa,
Alexey Skidanov, Sudhir Kumar.

Run began: Mon Nov 4 21:00:51 2024

Record Size 1 kB
File size set to 1048576 kB
Command line used: iozone -t1 -i0 -i2 -r1k -s1g .
Output is in kBytes/sec
Time Resolution = 0.000001 seconds.
Processor cache size set to 1024 kBytes.
Processor cache line size set to 32 bytes.
File stride size set to 17 * record size.
Throughput test with 1 process
Each process writes a 1048576 kByte file in 1 kByte records

Children see throughput for 1 initial writers = 248653.20 kB/sec
Parent sees throughput for 1 initial writers = 29222.11 kB/sec
Min throughput per process = 248653.20
kB/sec
Max throughput per process = 248653.20 kB/sec
Avg throughput per process = 248653.20 kB/sec
Min xfer = 1048576.00 kB

Children see throughput for 1 rewriters = 569351.94 kB/sec
Parent sees throughput for 1 rewriters = 25733.42 kB/sec
Min throughput per process = 569351.94
kB/sec
Max throughput per process = 569351.94 kB/sec
Avg throughput per process = 569351.94 kB/sec
Min xfer = 1048576.00 kB

Children see throughput for 1 random readers = 482187.00 kB/sec
Parent sees throughput for 1 random readers = 471858.58 kB/sec
Min throughput per process = 482187.00
kB/sec
Max throughput per process = 482187.00 kB/sec
Avg throughput per process = 482187.00 kB/sec
Min xfer = 1048576.00 kB

Children see throughput for 1 random writers = 335508.47 kB/sec
Parent sees throughput for 1 random writers = 23622.89 kB/sec
Min throughput per process = 335508.47
kB/sec
Max throughput per process = 335508.47 kB/sec
Avg throughput per process = 335508.47 kB/sec
Min xfer = 1048576.00 kB



iozone test complete.
ubuntu@ubuntu-2204-rpi5:/media/ubuntu/rootfs/tmp$ cd -
/home/ubuntu

# SSD
ubuntu@ubuntu-2204-rpi5:~$ iozone -t1 -i0 -i2 -r1k -s1g .
Iozone: Performance Test of File I/O
Version $Revision: 3.489 $
Compiled for 64 bit mode.
Build: linux

Contributors:William Norcott, Don Capps, Isom Crawford, Kirby
Collins
Al Slater, Scott Rhine, Mike Wisner, Ken Goss
Steve Landherr, Brad Smith, Mark Kelly, Dr. Alain CYR,
Randy Dunlap, Mark Montague, Dan Million, Gavin
Brebner,
Jean-Marc Zucconi, Jeff Blomberg, Benny Halevy,
Dave Boone,
Erik Habbinga, Kris Strecker, Walter Wong, Joshua
Root,
Fabrice Bacchella, Zhenghua Xue, Qin Li, Darren
Sawyer,
Vangel Bojaxhi, Ben England, Vikentsi Lapa,
Alexey Skidanov, Sudhir Kumar.

Run began: Mon Nov 4 21:03:30 2024

Record Size 1 kB
File size set to 1048576 kB
Command line used: iozone -t1 -i0 -i2 -r1k -s1g .
Output is in kBytes/sec
Time Resolution = 0.000001 seconds.
Processor cache size set to 1024 kBytes.
Processor cache line size set to 32 bytes.
File stride size set to 17 * record size.
Throughput test with 1 process
Each process writes a 1048576 kByte file in 1 kByte records

Children see throughput for 1 initial writers = 97677.51 kB/sec
Parent sees throughput for 1 initial writers = 84612.58 kB/sec
Min throughput per process = 97677.51
kB/sec
Max throughput per process = 97677.51 kB/sec
Avg throughput per process = 97677.51 kB/sec
Min xfer = 1048576.00 kB

Children see throughput for 1 rewriters = 130274.91 kB/sec
Parent sees throughput for 1 rewriters = 104275.54 kB/sec
Min throughput per process = 130274.91
kB/sec
Max throughput per process = 130274.91 kB/sec
Avg throughput per process = 130274.91 kB/sec
Min xfer = 1048576.00 kB

Children see throughput for 1 random readers = 432878.72 kB/sec
Parent sees throughput for 1 random readers = 431070.14 kB/sec
Min throughput per process = 432878.72
kB/sec
Max throughput per process = 432878.72 kB/sec
Avg throughput per process = 432878.72 kB/sec
Min xfer = 1048576.00 kB

Children see throughput for 1 random writers = 104523.77 kB/sec
Parent sees throughput for 1 random writers = 97122.94 kB/sec
Min throughput per process = 104523.77
kB/sec
Max throughput per process = 104523.77 kB/sec
Avg throughput per process = 104523.77 kB/sec
Min xfer = 1048576.00 kB



iozone test complete.

>> https://groups.google.com/d/msgid/hbrobotics/CA%2BKVXVPg5GLWUmFoz2GUFoZ-0G-3MGbg_0CZcTBpb64qUYOivw%40mail.gmail.com <https://groups.google.com/d/msgid/hbrobotics/CA%2BKVXVPg5GLWUmFoz2GUFoZ-0G-3MGbg_0CZcTBpb64qUYOivw%40mail.gmail.com?utm_medium=email&utm_source=footer>.

>
> --
> You received this message because you are subscribed to the Google
> Groups "HomeBrew Robotics Club" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to hbrobotics+...@googlegroups.com

> <mailto:hbrobotics+...@googlegroups.com>.

> To view this discussion visit

> https://groups.google.com/d/msgid/hbrobotics/996eba9f-5a03-4341-b1b3-39f35003c109%40hal-pc.org <https://groups.google.com/d/msgid/hbrobotics/996eba9f-5a03-4341-b1b3-39f35003c109%40hal-pc.org?utm_medium=email&utm_source=footer>.