On Thu, Nov 10, 2011 at 4:58 AM, William Hay <[email protected]> wrote:
>> Source tarball and binary packages will be available early next week.
> The scalablelogic.com website looks more than a bit unfinished.  Some
> details of the optional commercial support would be appreciated.

Hi William,

Yes the website is more than a bit unfinished, given that we only had
less than 2 weeks to register for a site, writing more formal release
notes, registering for trademarks (which is important), and then
joining ISV partnership programs, together with creating an official
release for SC11. (We skipped the incorporating step because it is
currently under the consulting company I own).

I decided to start the Scalable Grid Engine Support business only
recently - in fact 2 weeks ago. We (myself and other contractors) have
been in the Grid Engine supporting/consulting business for a long
time, but we've only worked with local companies and people that we
know in person (eg. software house and investment bank in Toronto, and
remotely a biotech company in Europe where a friend works). On the
mailing list, I worked with Ron Chen to improve the Sun Grid Engine
code even before I joined the original SGE "users" mailing list. So
that's basically 10+ years of contributing to the SGE code base &
offering free technical support to SGE users, no matter Grid Engine
was under Sun Microsystems or Oracle's control.


I am co-founding this business not because I think this is the right
time to charge SGE users money instead of offering it for free.
Instead, more and more users and companies in the Grid Engine
ecosystem approach me and complain about the state of things
(open-core, and bundling of proprietary products with the offering,
and expensive support options just to name a few) - this is not the
Grid Engine project everyone used to love when Sun was in control. As
there was only a few weeks from SC11, companies in the GE ecosystem
wanted a better story to tell their customers and potential customers.
So as a long term Grid Engine contributor and maintainer of the
official open source Grid Engine version, I've decided to step up and
rebuild the community.

I understand that it is hard to make money from open source, and I
believe my time could be spent on client sites or other work that can
make "easier money". However, if money were the main and only goal,
then I would not have contributed to Grid Engine since 2002 in the
first place.


I will announce my intent early next week, together with the rest of
the code for the Grid Engine 2011.11 release. But let me guarantee you
and everyone on the list: Open Grid Scheduler is an open source
project and will remain so. (Scalable Logic & Scalable Grid Engine
support is an optional technical support program.) While I don't have
a crystal ball and can't tell you what I will be doing next year, I
can guarantee you that there will be at least 1 release every year
(and I have signed up to contribute new Grid Engine features & code
for another 10 years), and Grid Engine will not be open-core when the
Open Grid Scheduler is under my control.

If I can't contribute time and/or energy to enhance Open Grid
Scheduler/Grid Engine, then instead of forking the Open Grid Scheduler
code, one of the seven other guys on the Open Grid Scheduler team can
take over the project and also use "Scalable Grid Engine" as the
trademark & service mark:

http://sourceforge.net/project/memberlist.php?group_id=343697


For now, Open Grid Scheduler is under my control, and will remain so
for a little while (at least 1 year I believe) not because I am a
dictator, but rather we need a gatekeeper who knows what changes need
to be rejected. I worked with IP lawyers (when I worked at a large
computer company) and in fact I just bumped into one at lunch
yesterday (nothing to do with Grid Engine, he is just a friend's
friend).

I don't actively look at other people's patents, but I do hope to at
least filter out the obvious danger, like I did for the Linux kernel:

http://lwn.net/Articles/359070/

      "Rayson Ho found that Apple (NeXT, actually) has patented the technologies
       behind universal binaries, as a "method and apparatus for
architecture independent
       executable files" (#5432937 and #5604905)."


As of now, the Open Grid Scheduler code base has code contributed by
Sun Microsystems (and imported into the project svn by Ron Chen), and
modifications from us (or contractors hired by us, so we own the IP),
and is relatively safe. Without a gatekeeper to handle the filtering,
Grid Engine would not be useful to other companies in the GE
ecosystem.

Rayson




>
> Just typed "open grid scheduler" into google (in order to get a link
> to the project page) and top of the results is an ad from Univa.
>>
>>
> We're currently still on 6.2u3 and obviously need to pick a successor
> version.  My current thinking on the non-technical advantages of the
> various Grid Engine Successors
> Oracle: Commercial support from a Really Big Company
> Univa: Commercial support by a company that employs most of the
> original grid engine developers.
> Open Grid Scheduler:  Commercial support available for a 100% open
> source product.
> Son of Grid Engine: 100% Open source project (modulo Fritz's comments
> re:documentation) plus a huge easily accessible bug/feature request
> tracker.
>
> Presumably the other versions  could clone Dave's bug database since
> he cloned it from the original SGE bug tracker and most of the bugs
> probably apply
> to  the other versions as well.
>
> William
>

_______________________________________________
users mailing list
[email protected]
https://gridengine.org/mailman/listinfo/users

Reply via email to