Hi Bhaskar:
Cloud computing is an expansive subject.
Is there already some area that you are specialized in/worked in
and like to utilize that knowledge ?
thanks
Saifi.
>
> Hello
>
> I am Graduate student having knowlegde on Distributed Computing,
> Hadoop and some knowledge on mapreduce programming.Now I am looking
> the way for Cloud Computing so I am looking suggestion
>
> On Apr 3, 12:05 pm, Saifi Khan <saifi.k...@twincling.org> wrote:
> > On Fri, 3 Apr 2009, Bhaskar wrote:
> >
> > > Hello everybody
> >
> > > Can you suggest me the possible areas on cloud computing for Master's
> > > thesis.My Background is Computer Engineering
> >
> > > Thanks
> >
> > Hi Bhaskar:
> >
> > Cloud computing is an expansive subject.
> >
> > Is there already some area that you are specialized in/worked in
> > and like to utilize that knowledge ?
> >
> > thanks
> > Saifi.
>
Here is one suggestion:
Relational Database as a service on Hadoop/EC2.
thanks
Saifi.
Hi Varadarajan:
It would be good if you could share your knowledge on the topic
of 'connection brokers'.
i'm also keen to know if there are any new design patterns you have
observed in connection management ?
The DB guys, the PROXY guys and the ORB guys were doing this
for quite sometime.
Have a nice day !
thanks
Saifi.
It would be good if you could share your knowledge on the topic
of 'connection brokers'.
>
> Virtualisation plays an important role in cloud computing.
>
> Well ! We virtualised the server and virtual machines are all sitting
> there but how to route an incoming connection to a specific machine in the
> cloud. I found that all are proprietary solutions based on VMware.
>
You make a very important point.
At Twincling we did think of the same during the engg team meetup
and came up with VirtuE idea. However beyond the logo and choice
of FreeBSD as a platform for implementation, it is currently vaporware
:(
http://www.twincling.org/virtue
thanks
Saifi.
in case somebody is interested in 'security aspects of cloud',
i'd like to recommend a security model based on 'artificial
immune system'.
In 2007, i did read a bit of Ivan Roit's 'Essential Immunology'.
Failure in distributed systems is well modelled around the lines
of infection in a biological system as we are multi-cellular.
There are some papers by Leandro and Timmis.
@Shameer any interest in this area ?
thanks
Saifi.
Most of the security architectures we look at are
. end-point
. end to end
In cloud computing terms, this means:
. end-point --> cloud perimeter
. end to end --> over my cloud
In the cloud model, a lot of traffic would be "inside the cloud"
This contrasts with the model, most brittle security companies
are used to looking at.
Hence my emphasis on 'multi-cellular system model'.
Anyway, the main websites (though not the best) are:
http://www.cloudsecurityalliance.org/
and
http://cloudsecurity.org/
The usual gokers are all listed up on these sites.
My next startup idea is 'legal hold in the cloud' :)
thanks
Saifi.