I was curious if anyone had ever thought about renting linux clusters for
rendering with PovRay.  Google and Amazon both rent servers for cloud computing.
 Amazon's service is called EC2, and I'm not sure what Google's is called.  I
remember reading that previous versions of Povray were compatible with PVM.  So
has anyone thought about doing this, whether it's possible, and how expensive it
would be?  I think Amazon's service is about 10 cents per server-hour so in that
case it might only cost a few bucks for a render that would be computationally
prohibitive on a single machine.

Post a reply to this message

> I was curious if anyone had ever thought about renting linux clusters for
> rendering with PovRay.  Google and Amazon both rent servers for cloud computing.
>   Amazon's service is called EC2, and I'm not sure what Google's is called.  I
> remember reading that previous versions of Povray were compatible with PVM.  So
> has anyone thought about doing this, whether it's possible, and how expensive it
> would be?  I think Amazon's service is about 10 cents per server-hour so in that
> case it might only cost a few bucks for a render that would be computationally
> prohibitive on a single machine.
>
>
>

Rendering on various clusters have been done for quite some time, and 
not only on Linux clusters. There was even a discontinued BOINC project 
dedicated to distributed POV-Ray rendering.

Using version 3.6.x, it was problematic when using radiosity, but now, 
with version 3.7, it should work much beter.



Alain

Post a reply to this message

I recently tried this on EC2. Here is a pricing link
http://aws.amazon.com/ec2/pricing/
The 'Cluster Eight Extra Large' at $2.4 per hour
  +Trace Time: 0 hours  5 minutes 49 seconds (349.450 seconds)
  +using 32 thread(s) with 1116.588 CPU-seconds total
The 'High-CPU Extra Large' at $0.66 per hour
  +Trace Time: 0 hours 15 minutes 28 seconds (928.374 seconds)
  +using 8 thread(s) with 3013.769 CPU-seconds total
At home i5 quad core 3.5Ghz (3.9 Turbo) desktop
  +Trace Time: 0 hours  7 minutes 48 seconds (468.474 seconds)
  +using 4 thread(s) with 901.348 CPU-seconds total

Using 'cat /proc/cpuinfo' shows that the virtual cores are:
Xeon E5506 @2.13GHz
cache size: 4096 KB

This performance is a little disappointing, but there is no reason more
instances cant be spooled up rendering different parts of a scene with things
like 'Start_Row', but that gets pricey. I might use this in the future for a
large project i dont want to run on my desktop, but it wont be to make the
render faster.

Nothing special is needed to run on these Ubuntu 12.04 boxes, this is the
command i ran to install pov 3.7 (if i missed some dependencies, config will
tell you what they are)
I've shortened the povray link to prevent line break issues in posting.
It links to
http://www.povray.org/redirect/www.povray.org/beta/source/povray-3.7.0.RC6.tar.gz
=====
sudo apt-get update
sudo apt-get install libtiff-tools g++ libtiff4-dev libjpeg-dev libpng++-dev
libpthread-stubs0-dev libboost-all-dev make
wget http://goo.gl/inBqf
..../configure
make install
=====

From there its all normal, though you get charged any time the box is running,
so dont 'write code' on it. Write at home and push to the box.
Also, the billing counts partial hours as full, so that 5 minute test render
cost $2.40, i could have doubled the resolution and gotten billed the same.

Hope this was informative

Post a reply to this message

I'm resurrecting this old thread because I wanted to share my experiences with
rendering on an EC2 cluster.

As a benchmark, I rendered a large scene on my 3.2Ghz quad core desktop and it
took about 36 hours.

The key to making EC2 cost-effective is to use spot instances, which means they
are way cheaper but you run the risk of losing them.  In my experience, however,
the prices stay stable for days at a time, so if you bid high, you should be
fine.

I used high-cpu instances, which consist of 8 not-so-impressive cpu cores and
they only cost me $.07 per server per hour.  Compare this to the $.66 per
server/hour for an on-demand instance and this is a real bargain.  The catch is
that the price is not static, and if it rises above your bid, you could lose
your instance.  So best practice is to make your application "interruption
tolerant" which I think should be pretty doable.  I didn't do that, but instead
set a very high bid and hoped I didn't get hosed.  I didn't.

I took out 8 of these instances and set my 64 cores to work on my render.  I
didn't use anything clever for task distribution.  I just divided my image
horizontally into 8 equally-spaced strips and sent each one to a core.

the breakdown for server-hours was:
2 nodes took 2 hours
3 nodes took 3 hours
3 nodes took 4 hours

for a total of 25 server hours and a total cost of $1.75

With some bash kung-fu I believe that the price and/or render time could
probably be optimized a little more, but as a first order approximation
 this gives an idea what a render should cost.  For me I can guess that for
every 24 hours it would take me to render on my machine, it will cost about
$1.20, with render time depending on my cluster architecture, but theoretically
could be made arbitrarily small (just watch out because Amazon won't charge for
partial hours).

One word of caution is that this render was particularly well-suited for this
type of task distribution.  By that I mean that the hard to render stuff is more
or less uniformly distributed throughout the image.  An image that had mostly
background with all the hard stuff in one place would require a little more
thought.  Otherwise you could end up with all your nodes except one or two
finishing early and the last ones taking forever.

I do believe that cloud rendering can make renders possible that wouldn't be
feasible otherwise or can free up your machine for other things.  And with some
bash scripting could be done at minimal cost.

I also think that when povray supports mpi natively, we'll all be better off :)

Post a reply to this message

On 11/26/2012 10:08 PM, taftj wrote:
> I'm resurrecting this old thread because I wanted to share my experiences with
> rendering on an EC2 cluster.
>

Ran across an Amazon Web Services blog example using POV-Ray from 19 Aug 
2015 at:

https://aws.amazon.com/blogs/compute/better-together-amazon-ecs-and-aws-lambda/

"Example Setup: Ray-tracing high-quality images in the cloud"

Bill P.

Post a reply to this message