[Bio-Linux] Cloud computing

[Dawn Field]

Hi Tony,

This is the 4th or 5th time cloud computing has come to my attention in the past months, so of interest and happy to hear more thoughts about it - it is on the cusp of becoming a reality for some people.

I should note that might comment about more people talking about cloud computing is coming from the 'big groups' who are outgrowing even very big supercomputers due to the growth in metagenomic/genomic data.  

We are already working with the National GRID Service to deal with the 'big jobs' that people might want to do via Bio-Linux, but for which the hardware doesn't suffice (e.g.  big all against all homology searches of many genomes/metagenomes for example).

Tasks done by individuals are  best done 'at home' as you say, and I agree, and this is what Bio-Linux is for and always will be - to enable more people to do their own analyses.

The original post was about trying to get Bio-Linux loaded into the cloud for yet more people to use - big or little jobs, I assume...the concept of having access to software from anywhere.  Still an interesting idea.  The issue is that you have to pay and academics aren't used to that (I don't want it at all).  Perhaps, for example, in the future a teaching lab could be willing to pay a small fee to give a group of students access to a bio-linux teaching lab 'via the ether' instead of taking the time/effort to install it locally.


Dawn

>>> rwt017 at abdn.ac.uk 12/3/08 12:25:03 >>>
Dawn Field wrote:
> Thanks for bringing up cloud computing.  Yes, there are academic
> groups starting to use cloud computing and we, as a community, should
> start to think about how the NERC community may/may not use cloud
> computing in general in the future.
> 
> Is this coming up on the radar for more people on this list?

Hello, Dawn.

I'm not in favour of 'cloud' computing, because I see it as an attempt 
to centralise resources and take control of resources away from the 
people who use them. The particular feature of NEBC that I admire and 
have followed in my recommendations to NuGO is the distributed resource 
you have deployed to NERC grant holders, and we have emulated in NuGO.

Although I now work in bioinformatics, I have many years experience of 
working as a biologist and, in my experience, if you take control of a 
resource away from biologists they will create their own. All too often 
this results in people using their own desktop/laptop because *they* 
control how it is used. The great virtue I see in NEBC's present 
strategy is that you *do* give control of Bio-Linux workstations to the 
biologists concerned.

The most significant change to Bio-Linux I made in deploying our NBX's 
(NuGO Black Boxes), apart from using Ubuntu, is that I've used 
server-grade PC's and use the NBX's as lab-scale servers. In that 
repect, I'm balancing the freedom of the lab to use it's own NBX server 
in the way they want to against the administrative burden of each 
biologist administering their own NBX workstation. This is not a long 
way from what you do with Bio-Linux anyway, and we hope to collaborate 
with NEBC on deployment of Bio-Linux5 on our 31 NBX's at NuGO partners.

I've now marked my 'biobuntu' blueprint as obsolete, and will use 
Bio-Linux5 on both the RINH/BioSS Beowulf and NuGO NBX servers.

Bye,

	Tony.
-- 
Dr. A.J.Travis, University of Aberdeen, Rowett Institute of Nutrition
and Health, Greenburn Road, Bucksburn, Aberdeen AB21 9SB, Scotland, UK
tel +44(0)1224 712751, fax +44(0)1224 716687, http://www.rowett.ac.uk 
mailto:a.travis at abdn.ac.uk, http://bioinformatics.rri.sari.ac.uk/~ajt