Dear All,
At the 1st XLDB workshop we agreed to meet again in
about a year, somewhere in SF Bay Area, likely at SLAC.
To make this happen, it is time to start planning
the 2nd XLDB workshop now.
Where/when
==========
I'm proposing to meet for 2 days on September 29th
and 30th at SLAC. Let me know if you are planning
to attend and this slot doesn't work for you (but
do realize we won't be able to accommodate everybody's
schedule)
Program
=======
Based on the feedback received last year, the style of
the 1st workshop worked well, so we will try to keep
it similar this year: lots of discussions and very few
presentations.
We (myself, K-T and few others I talked to) envision
this year's workshop to be a continuation of the
discussions we started last year, with the main focus
on topics like:
* complex analytics - commonalities between science
and industry
* scalability, fault tolerance and crash recovery of
extremely large databases
* common features needed (provenance?, uncertainty?,
time travel?...)
* resource allocation/management
* on the fly reconfiguration and on the fly data
repartitioning
* [... suggestions what else welcome...]
The goal would be to identify commonalities,
understand and prioritize needed features and
discuss path forward.
We are considering having a session about war stories
(LHC? PanSTARRS? EBay? others?) - if you have a large
production database setup and would like to talk about
your war stories related to extremely large databases,
please let me know.
We are also considering having a dedicated time slot
for new startups (and maybe well established vendors),
so...
1) If you represent a startup involved in providing
extremely large database solutions and would like to
speak about it, please let me know. The form of this
session is to-be-decided (poster session? panel?
short presentations?).
2) If you would like to see a slot dedicated to well
established DBMS vendors, please let me know. Based
on the feedback I received last year this seems to be
less desired.
Finally, we believe it'd be valuable to spend the second
half of the second day on a tutorial related to the new
open source database that was proposed at the Asilomar
meeting. See the report:
http://xldb.slac.stanford.edu/download/attachments/4784226/sciDB2008_report.pdf
for more details about this meeting. We will provide
more information in advance to help you decide if you
want to stay or not.
Organizing Committee
====================
I'm looking for 2-4 volunteers who would like to help
with organizing this workshop. An up-front warning:
each member of the organizing committee will be expected
to do some real work (announcements, preparation, maybe
moderating, etc). If you are interested, please respond
directly to me.
Invitations
===========
I can already tell there is huge interest in this workshop,
and it will be a challenge to keep it relatively small,
e.g. under 70. We need to come up with a policy what criteria
to use to select who should be invited. (Keep in mind that
being on this list and receiving this email does _not_ guarantee
getting an invite...). I think we should shoot for no more
than two representatives for "bigger" sites/labs/companies,
and one representative otherwise.
Website
========
In the short term, up-to-date page dedicated to workshop
preparation will be maintained at:
http://xldb.slac.stanford.edu/display/XLDB/xldb2
Eventually we will create an official web page similar to:
http://www-conf.slac.stanford.edu/xldb07
Comments welcome
================
If you have suggestions regarding agenda/topics to discuss,
or any other comments/suggestions, please contact me or
respond to this email.
Based on the feedback we received, last year's meeting
was a great success, which means it'll be a big challenge
to make the 2nd workshop even better. I am looking forward
to making this happen together with you.
Best regards,
Jacek
--
Jacek Becla
Stanford Linear Accelerator Center
[log in to unmask]
+1 650 926 8664
