It carries through here... This was brought up as a recommendation in Launchpad (here in this bug report) back in 2019, In that bug report, I questioned why this had been ignored, and not discussed much since then. It didn't go away, and it was discussed as it should have been. I was embarrassed that it had been that way for 4 years.
Since then:
By then Debian had already removed it from being installed as a default. Ubuntu kept it. even after that bug report...
RedHat had removed it from being default installed.
PopOS had removed it from their default installed.
SUSE, is a special case, where they kept it for their lineup (which includes their Enterprise Server Lineup and desktops. I confirmed)... BUT then on page 16 of their Performance Analysis, Tuning and Tools Guide (https://documentation.suse.com/sbp/server-linux/pdf/SBP-performance-tuning_en.pdf), that chapter starts out with this quote:
>>> A correct IRQ configuration – above all in multi-core architecture and multi-thread
>>> applications– can have a profound impact on throughput and latency performance
...and further says that the first step to get there is to disable irqbalance (where they give the instructions to disable the service) and how to go through irq configuration from there.
Additional to the blog article linked to in the last comment above, I found this blog (https://blogs.oracle.com/linux/post/irqbalance-design-and-internals), that goes into how it makes decisions in load balancing and is best summed up in it's conclusion:
>>> This article described the internals of the irqbalance daemon. The information provided
>>> here can be used to debug and better understand load balance decisions taken by irqbalance.
The question I have is, if Ubuntu is Debian Branch, and we long ago went from having different kernels for desktop & server in ubuntu-base, but do have ubuntu-server packages and ubuntu-desktop packages, where things could be different, why is this still a broad sweep as a default install "for all"?
I think the above weighs in on having it as optional. But I am not in that that (final) decision.
I am happy that this is getting discussed properly now so that we can relook at this, and what it means to us today.
I said my initial piece and recommendation here: /bugs.launchpad .net/ubuntu/ +source/ irqbalance/ +bug/2046470/ comments/ 2
https:/
It carries through here... This was brought up as a recommendation in Launchpad (here in this bug report) back in 2019, In that bug report, I questioned why this had been ignored, and not discussed much since then. It didn't go away, and it was discussed as it should have been. I was embarrassed that it had been that way for 4 years.
Since then:
By then Debian had already removed it from being installed as a default. Ubuntu kept it. even after that bug report...
RedHat had removed it from being default installed.
PopOS had removed it from their default installed.
SUSE, is a special case, where they kept it for their lineup (which includes their Enterprise Server Lineup and desktops. I confirmed)... BUT then on page 16 of their Performance Analysis, Tuning and Tools Guide (https:/ /documentation. suse.com/ sbp/server- linux/pdf/ SBP-performance -tuning_ en.pdf), that chapter starts out with this quote:
>>> A correct IRQ configuration – above all in multi-core architecture and multi-thread
>>> applications– can have a profound impact on throughput and latency performance
...and further says that the first step to get there is to disable irqbalance (where they give the instructions to disable the service) and how to go through irq configuration from there.
Applications vendors, which we have in our repo's, such as Valve Steam and CpuFreq, currently recommend removing irqbalance, if installed. /github. com/ValveSoftwa re/Proton/ issues/ 3243 konkor. github. io/cpufreq/ faq/
RE:
https:/
http://
Additional to the blog article linked to in the last comment above, I found this blog (https:/ /blogs. oracle. com/linux/ post/irqbalance -design- and-internals), that goes into how it makes decisions in load balancing and is best summed up in it's conclusion:
>>> This article described the internals of the irqbalance daemon. The information provided
>>> here can be used to debug and better understand load balance decisions taken by irqbalance.
The question I have is, if Ubuntu is Debian Branch, and we long ago went from having different kernels for desktop & server in ubuntu-base, but do have ubuntu-server packages and ubuntu-desktop packages, where things could be different, why is this still a broad sweep as a default install "for all"?
I think the above weighs in on having it as optional. But I am not in that that (final) decision.
I am happy that this is getting discussed properly now so that we can relook at this, and what it means to us today.