Author Topic: 2u Blackbird Build with 18 cores?!  (Read 18638 times)

deepblue

  • Newbie
  • *
  • Posts: 13
  • Karma: +3/-0
  • The only winning move is not to play...
    • View Profile
2u Blackbird Build with 18 cores?!
« on: April 25, 2020, 09:09:46 am »
I wanted to provide a 'trip report' on the last ~6 months using the Blackbird as a 2u server platform.

The goal was to have a server with a similar footprint to an 1u/2u SuperMicro x86 rack mount 'appliance'. I paired this with an 18 Core Power 9 CPU as the logistical issues with procuring 8 core CPUs at the time was enough to twist my arm into spending another $1000. I really wanted on this platform, logistics and TDPs be damned!

There are serious thermal concerns that need to be addressed when considering a case.I would not recommend buying a case that you are uncomfortable with modifying, or is not explicitly designed for CPUs with a TDP of over 200 watts. RaptorCS does not recommend paring Blackbird motherboards with CPUs that have a TDP of over 160 watts. Since this configuration removes all thermal safety barriers, I knew that moving airflow in a short path and attaching heat sinks to the VFR chips would be mandatory.

Here is a parts list of the server (with prices at the time of purchase):

Code: [Select]
Power9 - 18 Core CPU @ $1,570
Blackbird MicroATX Motherboard @ $999.00
2u Power9 Heatsink @ $80.00
2x 40x28mm High CFM fans @ $17.00
2x 32gb DDR4 ECC Memory @ $294.00
Rosewill 2u Server Chassis @ $85.00
2x Noctua 80mm case fan @ $20.00
Corsair CX Series 750w PSU @ $69.99
2x 256gb SanDisk 2.5” SSD Hard Drive @ $68.00
2x 2.5” to 3.5” Drive adapter @ ~ $10.00
6x 2tb Crucial 2.5” SSD Hard Drive @ $1337.88
ICY DOCK 6 x 2.5” SATA Cage @ $61.50
2x Noctua 40x10mm fans @ $28.00
LSI 9211-8I PCI-E SAS adapter @ $68.96
2x MiniSAS to SATA cable @ $24.00
Phanteks PWM Fan Hub Controller @ $19.99
Various cables and screws (included and acquired over the years)

The case I went with is the humble Rosewill 2U Server Chassis/Server Case/Rackmount Case https://www.amazon.com/gp/product/B01F7RJPHO/. At $85 it was a great deal for what it provided. I was looking at a few specifics like low cost, support for a standard ATX power supply, and a detachable/modular internal disk tray design. But the most important feature was the 5.25” drive bay, as I was planning on installing an ICY DOCK 6 x 2.5" SATA cage to allow easy disk access and a quick way to add disk density while allowing for maximum airflow through the case.



Since I chose to go with the the low cost route on this case, the lid required a modification. I drilled an 80mm hole into the corner near the CPU fan exhaust, which helps move hot stale air away from the CPU cooler and brought down temps about 20c.



I had removed the disk trays inside the chassis when I first unboxed the case, so there is quite a bit of room for air to flow freely. I am able to maintain an idle temperature of around 40 Celsius, and a max temperature of around 75 Celsius under artificial stress tests. The disks never go over 40 Celsius and the noise is... not bad!. I am using 2x 40mm high CFM fans that makes the sound of sitting inside an idle jet aircraft with headphones on. This however is fine, as I have it stored in a room that is not going to cause distractions for me in noise sensitive areas.



Not pictured are the heatsinks on the VFR chips, but I am using Raspberry Pi heat sinks that are linked below.

https://www.amazon.com/Easycargo-Heatsink-conductive-Regulators-8-8mmx8-8mmx5mm/dp/B079FQ22LK

I am still playing around with the idea of retrofitting a different case with improved airflow, but the way the CPU and RAM are aligned makes it a bit tricky to route cool -> hot airflow. I know that my CPU utilization is quite low, so your mileage may vary when it comes to thermals during real world tasks for long periods of time. All of the fans are centrally controlled with the Phanteks controller, as the motherboard controller is not good enough for what I am attempting.

I hope this helps others when considering 2u options for PowerISA servers. Let me know if I should try something out if it helps with keeping temps under control.

« Last Edit: April 25, 2020, 09:28:40 am by deepblue »

ClassicHasClass

  • Sr. Member
  • ****
  • Posts: 468
  • Karma: +36/-0
  • Talospace Earth Orbit
    • View Profile
    • Floodgap
Re: 2u Blackbird Build with 18 cores?!
« Reply #1 on: April 25, 2020, 01:22:27 pm »
That is indeed a pretty bleeding edge and I'm impressed by the lengths you went to. I'm leery of how well that would work under load, however (or for how long). What types of tasks is it doing? How does it perform if the CPU slices are loaded? Do you notice any throttling behaviour?

deepblue

  • Newbie
  • *
  • Posts: 13
  • Karma: +3/-0
  • The only winning move is not to play...
    • View Profile
Re: 2u Blackbird Build with 18 cores?!
« Reply #2 on: April 26, 2020, 08:35:20 am »
I did neglect to state that I am using the Indium pad that was provided with the 2u heat sink. It is a much required part to this setup.

I am running 4 containers running Golang/NPM/LAMP stacks with load balancing. Surprisingly, all the LAMP/Web based software is super lightweight and is noticeably quicker than when it ran on my old x86 server.

I am also running an x86_64 VM with Mikrotik's RouterOS, which was required as Mikrotik does not have a ppc version of RouterOS available on an .ISO image.

I have never been able to 'naturally' push the CPU over 10% utilization, so I am pretty sure that I am not even close to making a single CPU slice work hard and I do not see any throttling. I keep this in a air conditioned room that does not go over 72F, which definitely helps keep things under control.

I am constantly adding new services on the server, so hopefully I will have more real world numbers to report over the next few weeks and months.

Here are some temps when idle:



Here are some temps during a ~10 minute artificial stress test:


« Last Edit: April 26, 2020, 08:45:29 am by deepblue »

surf

  • Newbie
  • *
  • Posts: 25
  • Karma: +1/-0
    • View Profile
Re: 2u Blackbird Build with 18 cores?!
« Reply #3 on: April 28, 2020, 06:33:46 pm »
Thanks for sharing the details of your build!  Which CPU are you using?  The datasheet describes a 130w 18-core version and a 190w version.  It seems like the 130w CPU should not stress the blackbird board.  In your stress test, could you tell if the speed was 'throttled' as it got hot?


deepblue

  • Newbie
  • *
  • Posts: 13
  • Karma: +3/-0
  • The only winning move is not to play...
    • View Profile
Re: 2u Blackbird Build with 18 cores?!
« Reply #4 on: April 28, 2020, 07:53:56 pm »
I am using the 18c chip that RaptorCS supplies:

https://www.raptorcs.com/content/CP9M06/intro.html

I could not tell if the chip was throttled during the test. I did try to load of a few web pages from the container that is hosting my website, but there was no noticeable difference in usability during the test. It was an artificial stress test so I know that in this particular instance it was being nice to the rest of the system.
« Last Edit: April 28, 2020, 07:58:09 pm by deepblue »

tle

  • Sr. Member
  • ****
  • Posts: 466
  • Karma: +53/-0
    • View Profile
    • Trung's Personal Website
Re: 2u Blackbird Build with 18 cores?!
« Reply #5 on: April 29, 2020, 02:45:53 am »
Nice server. Great thanks for sharing your experience with everyone. I think mine is due soon too.

Seems to me your 18-core is the 130W variant CPU.

Regarding the CPU thermal, I think your 2U HSF might not be enough to cool the CPU on full load. I am using the stock Raptor 3U HSF on the 8 core POWER9 (02CY649) and at times the temperature reaches to 80C on full load.
Faithful Linux enthusiast

My Raptor Blackbird

deepblue

  • Newbie
  • *
  • Posts: 13
  • Karma: +3/-0
  • The only winning move is not to play...
    • View Profile
Re: 2u Blackbird Build with 18 cores?!
« Reply #6 on: April 29, 2020, 04:17:24 pm »
The two 40mm fans strapped to the heat sink are pushing quite a bit of air. There is positive airflow moving into the case with an exit out the lid where I drilled the fan port. Its unlikely for hot air to linger near the heat sink. If this were in a non-ventilated environment, it would heat the space up very quickly.
 
Is there a definitive test that can determine if there is throttling at high temp loads? I wouldn't know because of how low my temps and actual utilization are.

I am using the model below. It looks like its the 190w variant:



https://en.wikichip.org/wiki/ibm/power/02cy489

MPC7500

  • Hero Member
  • *****
  • Posts: 592
  • Karma: +41/-1
    • View Profile
    • Twitter
Re: 2u Blackbird Build with 18 cores?!
« Reply #7 on: April 29, 2020, 07:16:10 pm »
From my experience I can tell, that a 4-core even clock 3.8GHz at 110°C/ 230°F. I don't know if this behavior is normal. But that's it.

ClassicHasClass

  • Sr. Member
  • ****
  • Posts: 468
  • Karma: +36/-0
  • Talospace Earth Orbit
    • View Profile
    • Floodgap
Re: 2u Blackbird Build with 18 cores?!
« Reply #8 on: April 29, 2020, 10:14:38 pm »
Quote
It looks like its the 190w variant:

Yes, 02CY489 is a DD2.2 18-core @190W.

Borley

  • Full Member
  • ***
  • Posts: 177
  • Karma: +16/-0
    • View Profile
Re: 2u Blackbird Build with 18 cores?!
« Reply #9 on: May 05, 2020, 07:14:26 am »
That's a familiar looking CPU fan solution  ;)

If you ever do push it past 10% CPU usage, I would be interested to see if it doesn't shutdown on Blackbird's power contraints.

deepblue

  • Newbie
  • *
  • Posts: 13
  • Karma: +3/-0
  • The only winning move is not to play...
    • View Profile
Re: 2u Blackbird Build with 18 cores?!
« Reply #10 on: November 02, 2021, 08:42:38 am »
I wanted to give everyone an update on this particular setup, as its been a little over 2 years since I purchased this from RaptorCS.

I have been able to push the server beyond its natural 1-5% utilization, as backups definitely stress the CPU during an xz compression then saving it to a CIFS share. The Blackbird seems to handle the additional power draw just fine, and it operates without complianing about the CPU being a bit toasty (60c-80c on average).

Note: I am only allowing 18 threads during the xz compression as letting all 72 threads run gives you no performance benefit. If anyone has any ideas as to why this is, I would appreciate the knowledge.

Thermals have not really changed since adding the heat sinks to the VFRs with thermal epoxy. In response to this, I have set the max power utilization to 160w via the WebUI side of the BMC. This has seemed to result in very little perf loss and it makes me feel better knowing I wont blow out my VFRs on the Blackbird.

I have updated from Ubuntu 18.04 to 20.04 with a little hassle, as ZFS did not mount properly due to renamed disk paths. I am still rebuilding from bungled backups, so make sure that your backups are solid before doing anything drastic! Other than having to reclaim my Wordpress MySQL DBs from my backups, it was successful migration.

Other than the server being a bit 'lopsided' when it comes to CPU count/Memory, it has been a solid performer for all my personal web services and various linux server-y things. I would not hesistate to purchase another setup like this so I can have some redundancy, but that will have to wait due to costs and availability.

However, I am very excited to see what comes next from RaptorCS and I am closely watching the developments around Kestrel and their SoftBMC project. I will most likely repurpose this for a workstation machine after Raptor releases a board that has their latest and greatest kit, as there are definite reasons (for me) to move once the hardware drops.

ejfluhr

  • Newbie
  • *
  • Posts: 44
  • Karma: +3/-0
    • View Profile
Re: 2u Blackbird Build with 18 cores?!
« Reply #11 on: November 12, 2021, 11:42:44 pm »
Nice update...tx!  Do you know if the xz compression uses the in-processor compression accelerator, or does it just run on the cores?

Do you know if WOF (dynamic frequency boosting) is active?   Since you are so power limited, perhaps it is lowering the core frequency when you add more threads than 1 per core?


deepblue

  • Newbie
  • *
  • Posts: 13
  • Karma: +3/-0
  • The only winning move is not to play...
    • View Profile
Re: 2u Blackbird Build with 18 cores?!
« Reply #12 on: November 21, 2021, 11:42:21 pm »
Nice update...tx!  Do you know if the xz compression uses the in-processor compression accelerator, or does it just run on the cores?

It looks like it runs on the cores. I have to be careful with how many threads my xz compression consumes as it bogs down the system after I allocate 1 thread for every core. I also do gzip backups with mysql and that seems to not bother the CPU much with 18 threads.

Quote
Do you know if WOF (dynamic frequency boosting) is active?   Since you are so power limited, perhaps it is lowering the core frequency when you add more threads than 1 per core?

I am not sure if it is, but I am assuming so. The only real change that I have made is that I use a developer build of the blackbird firmware, as I needed help with unbricking it after a failed BIOS update. Maybe one of the folks here can determine if its active from the versions below?

BMC Image: 2.7.0-dev-581-g18878e4f6
Server Image: open-power-blackbird-v2.3-rc2-65-g1bd4a042-dirty

I am not sure if setting my max power draw to 160w is helping the situation, but I have upgraded the chassis fans to provide a lot more airflow to the CPU heatsink and VFRs.
« Last Edit: November 21, 2021, 11:55:45 pm by deepblue »

MauryG5

  • Hero Member
  • *****
  • Posts: 773
  • Karma: +22/-1
    • View Profile
Re: 2u Blackbird Build with 18 cores?!
« Reply #13 on: November 22, 2021, 01:20:55 am »
A good way to test the CPU can be to do some compilations and also be able to establish the number of cores or thread to use, I would have done some tests in the tel sense I think ...

ejfluhr

  • Newbie
  • *
  • Posts: 44
  • Karma: +3/-0
    • View Profile
Re: 2u Blackbird Build with 18 cores?!
« Reply #14 on: December 11, 2021, 12:25:55 am »
Re: the xz compression behavior, can you tell if you are memory bandwidth starved?  Lots of cores vs. Blackbird's meager DIMM capacity isn't a very balanced combination. 

BTW, I looked at your awesome graphs up above more carefully and you can see how current & voltage are moving around --> that looks like the processor is indeed dynamically boosting/dropping frequency to manage within it's power limit.