[Beowulf] Digital Image Processing via HPC/Cluster/Beowulf - Basics

Vincent Diepeveen diep at xs4all.nl
Sat Nov 3 06:09:50 PDT 2012

hi CJ,

I assume we speak about some thousands of images of artworks that  
need to get processed to recognize fraud
and avoid arttheft.

Now everything that has to do with those expensive artworks, usually  
big cash gets asked there.
Be careful there.

Just objectively seen this is not a difficult workload and requires  
relative little calculations and a first year student
so to speak would be able to handle this on an advanced mobile phone.

This is just about having good software that can do what you want.

Realize good image software on a videocard, say a nvidia tesla, can  
proces hundreds of photos a second.

So if all you have is just a few thousand of photos, you don't want  
to overspend upon buying hardware that can proces
everything your gallery has within 10 seconds.

Using a simple $200 computer that requires a few hours to do this is  
more than sufficient as for this type of image recognition
you will have plenty of time.

Your only consideration therefore is not whether you need a cluster  
or anything like that,
your only worry is avoiding your $200 computer getting hacked by  

Now THAT is a big problem, as in itself frauding with photos, say  
REMOVING from your database photos of artworks
they intend to steal, that's something you DEFINITELY want to avoid.  
As criminals can make big cash with this - assume
they also have military type software and equipment to carry out the  
online burglary and break in there. It's peanuts to
hire ex-NSA guys to do that.

Protecting against THAT is not easy.

So those security concerns will be 99.9% of your worry.

On Nov 1, 2012, at 7:31 AM, CJ O'Reilly wrote:

> Hello, I hope that this is a suitable place to ask this, if not, I  
> would equally appreciate some advice on where to look in lue of  
> answers to my questions:
> You may guess that I'm very new to this subject.
> I am currently researching the feasibility and process of  
> establishing a relatively small HPC cluster to speed up the  
> processing of large amounts of digital images.
> After looking at a few HPC computing software solutions listed on  
> the Wikipedia comparison of cluster software page ( http:// 
> en.wikipedia.org/wiki/Comparison_of_cluster_software ) I still have  
> only a rough understanding of how the whole system works.
> I have a few questions:
> 1. Do programs you wish to use via HPC platforms need to be written  
> to support HPC, and further, to support specific middleware using  
> parallel programming or something like that?
> OR
> Can you run any program on top of the HPC cluster and have it's  
> workload effectively distributed? --> How can this be done?
> 2. For something like digital image processing, where a huge amount  
> of relatively large images (14MB each) are being processed, will  
> network speed, or processing power be more of a limiting factor? Or  
> would a gigabit network suffice?
> 3. For a relatively easy HPC platform what would you recommend?
> Again, I hope this is an ok place to ask such a question, if not  
> please help refer me to a more suitable source.
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org sponsored by Penguin  
> Computing
> To change your subscription (digest mode or unsubscribe) visit  
> http://www.beowulf.org/mailman/listinfo/beowulf

More information about the Beowulf mailing list