Cloud Computing Breakout
John Brooke
Name of Chair: Steven McGough
Name of Scribe: John Brooke
What are the five most important things learnt during this discussion:
1. We have fifteen years of experience with distributed computing in a
Grid context and this should not be thrown away. Lessons of
authorisation and federation are particularly relevant.
2. Financial issues, e.g credit card payment. Academic institutions
dont like to see money go out of the institution to external agencies.
Issues: how credit card payments are accounted against projects.
3. Why are people using Cloud? Convenience, immediate access for cash,
attractive for people who needs computing in bursts. May not be so
good for heavy users. Balance of private/public provision of computing
resource.
4. IPR issues are very important, how much control do you have over
your data in the Cloud?.
5. Software as a service for academically produced software is an
attractive idea. Dont have to have multiple of software for different
OS's, lower entry point for users. E.g NeuroDebian, BioLinux.
What are the problems, and are there solutions?
1. Data costs are a major problem with certain applications, e.g. Next
Generation Sequencing produces huge amounts of data.
2. Internet connections do not have sufficient capacity for the data
transfer. Even in Janet MAN providers restrict access outside their
area, example Huddersfield trying to use a resource on the North West
Grid.
3. Cloud resources need to be located in the UK due to legal
requirements. E.g. medical and social science that have sensitive
data. Researchers want to know where their data physically resides.
What further work could be done, and who should do it?
Talk to Janet. Providers may be talking. Need representation on behalf
of the academic community, who should do this, SSI, NGS?
Funders and Universities need to be made aware of Cloud Computing and
how it fits with allocation and management of research funds. Cloud is
a recurrent not a capital expense.
Standardisation of Cloud APIs, work in various bodies e.g. Open Grid
Forum in Europe, NSF in US. Possible support of OGF standards, e.g.
BES.
Are there any useful resources that people should know about?
Cloud Providers now will invoice rather than require credit card payments.
Who knows about the Janet Cloud? How is it being made available to researchers?
Neil Chue Hong (SSI)
making their software available to other researchers without having to
invest in capital to purchase servers.
cheers,
neil
> --
> You received this message because you are subscribed to the Google
> Groups "Collaborations Workshop 2012" group.
> To post to this group, send email to coll...@googlegroups.com
> To unsubscribe from this group, send email to
> collabw12+...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/collabw12?hl=en?hl=en
--
Neil Chue Hong
Director, Software Sustainability Institute
EPCC, University of Edinburgh, JCMB, Edinburgh, EH9 3JZ, UK
Tel: +44 (0)131 650 5957
http://www.software.ac.uk/
LinkedIn: http://uk.linkedin.com/in/neilchuehong
Twitter: http://twitter.com/npch