Hi Michael,
I suppose the Beehive works with Amazon instances properly, considering you
install a data server on the cloud and configure it to connect to your
local metadata server (or else push the metadata server to the cloud and
configure your local backup machines to communicate properly).
About the LAN I presume this is just a latency requirement, to allow for
transparent use of the storage infrastructure as if it were local. If you
can cope with longer waiting times (as in case of backing data up) I don't
see any drawbacks to set your infrastructure over a WAN.
About CPU sharing, BeeFS is integrated with Hadoop so you can run MapReduce
jobs on your Hadoop cluster using BeeFS as the storage infrastructure. The
following document may be of some help:
http://redmine.lsd.ufcg.edu.br/projects/beefs/wiki/Using_BeeFS_With_Hadoop
Cheers,
Renato
2012/7/18 Michael Jonker <michael at openpoint.ie>
> Thanks Renato,
>> That is exactly the kind of thing that I am looking for, except suitable
> for 'cloud' usage (not LAN only). Something like p2p distributed cpu
> sharing, except for data storage. This is what Symform <
>http://www.symform.com/> is doing, except that they are proprietary, no
> Linux and expensive.
>> Beehive may (theoretically) be re-purposed to this, but that still leaves
> you with the task of building your own community of 'sharers'. What would
> be ideal is for a trusted FOSS vendor such as 'Ubuntu One' to make this
> typology where you can pay for storage with cash or disk space trading.
>http://www.infinit.io/ seem to be heading in this direction, but they are
> still in incubation.
>>> On 18/07/2012 12:16, Renato Miceli wrote:
>> Hi Michael,
>> I recommend the Beehive File System:
>http://www.lsd.ufcg.edu.br/beefs>> It is a distributed file system that harnesses the free disk space on
> machines (ideally in a LAN) to build a large, scalable, reliable and
> transparent (POSIX-compliant) storage infrastructure. Plus it can be used
> to back files up across different machines in the network (full and
> differential backups, scheduled backup, variable no. of replicas, affinity,
> etc).
>> Cheers,
> Renato
>>> 2012/7/18 Michael Jonker <michael at openpoint.ie>
>>> I am researching strategies for decentralised backup solutions which are
>> economical, scalable, suitable for SMB's and Linux compatible. Ideally
>> these should also be FOSS.
>>>> So far the only mature looking service I have found is Symform <
>>http://www.symform.com/>, but this is proprietary, has no Linux client
>> and IMO expensive if you are giving back equal/more disk space.
>>http://www.symform.com/>>>> I have also found these (not mature) options:
>> Infinit:
>>http://en.wikipedia.org/wiki/Infinit>>http://www.infinit.io/>>>> Backupp2p:
>>http://backupp2p.pbworks.com/w/page/11416763/FrontPage>>>> Does anybody have insight, advice, suggestions regarding distributed p2p
>> backup solutions?
>> I find the concept intriguing and progressive when compared to big data
>> warehouses.
>>>> Cheers,
>>>> Michael Jonker
>> --
>> Irish Linux Users' Group mailing list
>> About this list : http://mail.linux.ie/mailman/listinfo/ilug>> Who we are : http://www.linux.ie/>> Where we are : http://www.linux.ie/map/>>>>>>
Maintained by the ILUG website team. The aim of Linux.ie is to
support and help commercial and private users of Linux in Ireland. You can
display ILUG news in your own webpages, read backend
information to find out how. Networking services kindly provided by HEAnet, server kindly donated by
Dell. Linux is a trademark of Linus Torvalds,
used with permission. No penguins were harmed in the production or maintenance
of this highly praised website. Looking for the
Indian Linux Users' Group? Try here. If you've read all this and aren't a lawyer: you should be!
