[one-ecosystem] Exploring BlobSeer and GloBeM in the context of OpenNebula

[Bogdan Nicolae]

Dear all,

As a follow up to the discussions that Maria's group had with the OpenNebula
team,
here are a few details about BlobSeer.

BlobSeer is a large-scale, distributed storage service specifically designed
to offer
versioning support while delivering a high throughput for concurrent, fine
grain R/W
accesses to the same binary large object (BLOB) (presentation attached).

We recently found a series of interesting applications in the context of
clouds:

(1) using the versioning support of BlobSeer
to efficiently take snapshots of virtual machine images transparently,
without
relying on custom image file formats that support versioning (such as
QCOW2).

(2) using BlobSeer as an efficient data sharing solution for large-scale
applications
running inside VMs (instead of more conventional approaches such as Hadoop
Distributed
File System).

Below are some more references.

Best regards,
----
Bogdan Nicolae
University of Rennes 1, KerData Team, INRIA Rennes Bretagne-Atlantique
Tel: +33 (0) 2 99 84 22 78
Web: http://www.irisa.fr/kerdata/people/Bogdan.Nicolae


Links:
- KerData team's website:
http://www.irisa.fr/kerdata/<http://www.irisa.fr/kerdata/people/Gabriel.Antoniu/>
- The BlobSeer web page : http://blobseer.gforge.inria.fr/

Recent papers:

    1. The BlobSeer reference paper (approach, design principles,
algorithms):

Nicolae Bogdan ; Antoniu Gabriel ; Bougé Luc ; Moise Diana ; Carpen-Amarie
Alexandra
 <http://www.irisa.fr/kerdata/people/Alexandra.Carpen-Amarie>
 <http://www.irisa.fr/kerdata/people/Alexandra.Carpen-Amarie> BlobSeer: Next
Generation Data Management for Large Scale
Infrastructures<http://hal.inria.fr/inria-00511414>
*Journal of Parallel and Distributed Computing* (Aug. 24, 2010)
    2. Using BlobSeer to improve Hadoop:

Nicolae Bogdan ; Moise Diana ; Antoniu Gabriel ; Bougé Luc ; Dorier
Matthieu
 <http://perso.eleves.bretagne.ens-cachan.fr/~mdori307/>
 <http://perso.eleves.bretagne.ens-cachan.fr/~mdori307/> BlobSeer: Bringing
High Throughput under Heavy Concurrency to Hadoop Map-Reduce
Applications<http://hal.inria.fr/inria-00456801>
In *24th IEEE International Parallel and Distributed Processing Symposium
(IPDPS 2010)*, Apr. 21, 2010, Atlanta, United States
    3. Further improvements to Hadoop thanks to BlobSeer:

Moise Diana; Antoniu Gabriel ; Bougé Luc
 <http://perso.bretagne.ens-cachan.fr/~bouge/>
 <http://perso.bretagne.ens-cachan.fr/~bouge/> Improving the Hadoop
Map/Reduce Framework to Support Concurrent Appends through the BlobSeer BLOB
management system <http://hal.inria.fr/inria-00476861> In *The First
International Workshop on MapReduce and its Applications (MAPREDUCE'10)*,
Jun. 22, 2010, Chicago, United States

 4. BlobSeer as a QoS enabled storage service on the cloud for application
data
Montes Jesús; Nicolae Bogdan; Antoniu Gabriel; Sánchez Alberto; Pérez
Maria Using
Global Behavior Modeling to Improve QoS in Cloud Data Storage
Services<http://hal.inria.fr/inria-00527650/en>
In *CloudCom '10: Proc. 2nd IEEE International Conference on Cloud Computing
Technology and Science*, Nov. 30, 2010, Indianapolis, United States
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opennebula.org/pipermail/ecosystem-opennebula.org/attachments/20101202/1b52e624/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: slides.pdf
Type: application/pdf
Size: 4609817 bytes
Desc: not available
URL: <http://lists.opennebula.org/pipermail/ecosystem-opennebula.org/attachments/20101202/1b52e624/slides-0001.pdf>