Anyone interested in running a remote renderer process?

[Demetrios Tsillas]

Hi folks,

I'm speculating that it may be possible to have mojo run in a mode where it will copy all its data across its IPC socket (rather than use shared memory buffers). If this was possible and if it is was also possible to have mojo run over TLS or ws://, and if its also possible to remotely execute a compatible renderer process and have it's IPC sockets somehow bind to the locally executing browser (I know this is a lot but thanks for bearing with me) what would prevent running the renderer process on a remote system?

You may ask why this would be useful or practical but I'm envisioning remote rendering service which can provide a way to allow the renderer to do its drawing into some shared framebuffer which would appear on the local view.

Pure speculation but I wanted to ask for people's thoughts on this.

Thanks!!!

-Jim.

[Stefan Zager]

In theory, I think that would work. Currently, there are still non-mojo IPC channels between the browser and renderer processes, although we're in the process of converting them to mojo. Keep in mind there is also a mojo channel between the renderer and the viz process, which in your case would presumably be running on the same host as browser process.

--
--
Chromium Developers mailing list: chromi...@chromium.org
View archives, change email options, or unsubscribe:
http://groups.google.com/a/chromium.org/group/chromium-dev
---
You received this message because you are subscribed to the Google Groups "Chromium-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to chromium-dev...@chromium.org.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/chromium-dev/19613c4a-1355-4136-be4d-76cf7c336a13n%40chromium.org.

[Chris Hamilton]

My Google-fu is failing me, but there was a BlinkOn presentation a few years ago where somebody had basically done just that. They had modified Chrome so that all browser <-> renderer communication could happen over the network, and thus could be running on separate machines.

Cheers,

Chris

To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/chromium-dev/CAHOQ7J8fpdigEANArVdHKYR%3D%2B1xSbgyXVUS9W-xSGhyMScjq7A%40mail.gmail.com.

[Jeremy Roman]

Anything that uses shared memory expressly, passes file descriptors (I think some font stuff does this), expects performance timers to match, etc. also seems like it would be a massive pain to support, though not in principle impossible.

To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/chromium-dev/CAA2pCLdKMGwEg5MrmoYaHDquqAVREDRxibAWk%3DPHNpKyJeq40w%40mail.gmail.com.

[Demetrios Tsillas]

Wow if you can provide a link that presentation it would be awesome.

[Alexander Dunaev]

Hi Demetrios,
That presentation was on BlinkOn 10, its full name was "Chromium
Offloading Architecture and Technique using In-Home Edge Devices". The
presenters were from Samsung, and their idea was to delegate
CPU-expensive things like rendering or Javascript from low power devices
(such as smart watch) to something more powerful and connected to the
power source, like a TV or laptop.

They had a working demo and some statistics data that suggested that the
idea is worthwhile.

Unfortunately, I was unable to find a video recording, and actually I am
not sure that there was one.

>> [1].

>>
>> --
>> --
>> Chromium Developers mailing list: chromi...@chromium.org
>> View archives, change email options, or unsubscribe:
>> http://groups.google.com/a/chromium.org/group/chromium-dev
>> ---
>> You received this message because you are subscribed to the Google
>> Groups "Chromium-dev" group.
>> To unsubscribe from this group and stop receiving emails from it,
>> send an email to chromium-dev...@chromium.org.
>
>> To view this discussion on the web visit
>>

> https://groups.google.com/a/chromium.org/d/msgid/chromium-dev/CAHOQ7J8fpdigEANArVdHKYR%3D%2B1xSbgyXVUS9W-xSGhyMScjq7A%40mail.gmail.com
>> [2].

>
> --
> --
> Chromium Developers mailing list: chromi...@chromium.org
> View archives, change email options, or unsubscribe:
> http://groups.google.com/a/chromium.org/group/chromium-dev
> ---
> You received this message because you are subscribed to the Google
> Groups "Chromium-dev" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to chromium-dev...@chromium.org.
> To view this discussion on the web visit

> https://groups.google.com/a/chromium.org/d/msgid/chromium-dev/7be6eba2-bd0b-42b0-be0f-d6b1ec09721bn%40chromium.org
> [3].
>
>
> Links:
> ------
> [1] https://groups.google.com/a/chromium.org/d/msgid/chromium-dev/19613c4a-1355-4136-be4d-76cf7c336a13n%40chromium.org?utm_medium=email&amp;utm_source=footer
> [2] https://groups.google.com/a/chromium.org/d/msgid/chromium-dev/CAHOQ7J8fpdigEANArVdHKYR%3D%2B1xSbgyXVUS9W-xSGhyMScjq7A%40mail.gmail.com?utm_medium=email&amp;utm_source=footer
> [3] https://groups.google.com/a/chromium.org/d/msgid/chromium-dev/7be6eba2-bd0b-42b0-be0f-d6b1ec09721bn%40chromium.org?utm_medium=email&utm_source=footer

--
Regards,
Alex

[Paul Sery]

I'm interest but for another reason. I'd like to see if the browser process can connect with remote render processes to isolate to further secure the internal network.

I've not found related work since this post and would be appreciate any additional information or suggestions.

Thanks!

Paul