[tsc] [msc] Candidate Draft - Landscape White Paper
David Snelling
David.Snelling at UK.Fujitsu.com
Wed Mar 28 12:14:45 CDT 2007
Folks,
TSC met today the the aim of aligning the TSR and the LS papers. We
did the following.
1) Since the segmentation was the same in both papers, but in a
different order, we reordered the TSR to match the current layout in
the LS. This was easy, and had little impact on TSR. The segmentation
probably still makes some people unhappy, but at least the two
documents now agree.
2) With respect the definition of Grid, we followed Joel suggestion
plus some word smithing. The result is below. We suggest that the LS
document follow this approach, as it includes what they were
focusing on and is very consistent with what we had intended in the TSR.
============ New TRS Text ==============
The long-term vision of Grid can be summed up as follows: “Scalable
distributed computing across multiple heterogeneous platforms,
locations, and organizations.” The notion of distributed computing as
used in this definition includes a wealth of highly complex
technologies, some still the focus of research. This definition
includes the degenerate case of a homogeneous HPC cluster. It also
addresses operation across a wide area and possibly multiple domains
of administrative control. The security, privacy, economic, and
political aspects of Grids increase by orders of magnitude with the
introduction of Internet scale operation.
======================================
3) The TSC has been asked to read the LS document and provide feed
back to Mark and John. Please do so.
To this end, I would like to add my support to Jay's view on
virtualization. I would like to see this worked into the LS document.
If you can't make it fit, we might like to put in in the TSR.
For reference, Jay's comments:
This is pretty good but I normally make a distinction here.
Virtualization is typically thought of in terms of virtualizing IT
resources like computers, storage or networks. Grid takes the
virtualization concept to another level and actually virtualizes the
application and information resources that rely on the IT
infrastructure. As you said that de-couples the applications and
information from the IT infrastructure and allows the application and
information to use whatever infrastructure is there. At a certain
level the IT infrastructure is abstracted and made to appear
homogeneous regardless of wether is dynamic (virtualized) or not. It
is important to note that many Grids today are built mostly from non-
virtualizaed IT resources (e.g. computers, storage and networks that
are NOT virtualized) and yet the virtualization of the application
and the information still allows the Grid to create an environment
where resources can be pooled shared and re-purposed easily. When
you combine this type of Grid virtualization and the IT resource
virtualization the effects are multiplied even more. Now the static-
binding is broken both between application / information and
infrastructure but also between IT resources and physical hardware/
software.
4) Meanwhile, with respect to further work on the TSR, I have the
pen. I will, but have no idea when, work through the public comments
and either
a) Make the recommended change if it seems simple enough to do
without extended discussion, or
b) Post a tracker for discussion wrt the next version.
Then we will have a call on April 11 or 25 to agree my changes. If
anyone else would like the pen for a few days PLEASE let me know and
you can have it. NB The pen is currently available and the Public
comments can be found here [https://forge.gridforum.org/sf/discussion/
do/listTopics/projects.ggf-editor/
discussion.info_technical_strategy_document] if anyone is interested
this week.
Thanks and talk later.
--
Take care:
Dr. David Snelling < David . Snelling . UK . Fujitsu . com >
Fujitsu Laboratories of Europe Limited
Hayes Park Central
Hayes End Road
Hayes, Middlesex UB4 8FE
Reg. No. 4153469
+44-208-606-4649 (Office)
+44-7768-807526 (Mobile)
More information about the tsc
mailing list