[Ofa_boardplus] Logo Program Discussion

Fields, Parks M parks at lanl.gov
Fri Sep 1 10:05:33 PDT 2017


+1




parks
parks at lanl.gov<mailto:parks at lanl.gov>



On Sep 1, 2017, at 5:07 AM, Stephen Poole <swpoole at gmail.com<mailto:swpoole at gmail.com>> wrote:

So, I did not receive a copy of the slides, but I will pose some overall questions/comments to OFA.

Who do you think your customer base is?
What is your value add to your customer base?
How is this value perceived by your customer base?
Do you have buy-in from your customer base? (Obvious by their participation/commitment)
What is your "product"?
Why do I want your product?
Who supports your product?
Will it run on all of my HW/SW?
Do you have a support model?
Is it vendor supported?
Does it have a strong/vibrant community behind it?
Do I have a good "brand-name?"
How rapidly can you adapt to new HW offerings?

If OFA considers themselves the guardian of an API/Spec. they why would they worry about
a "plug-fest" or funding it? You HAVE to be vendor neutral. If you are not, then you should
either disband or rename yourself to "vendor-Y-Spec".

If you look at something like MPI. They have a spec/man pages/API... There are several
implementations. They do actually have V&V capabilities and a very vibrant community.
The last I knew, they did not "fund" directly a plug-fest. The users sort this out by not
using things that are not compatible.

What/how is OFA going to address things like CCIX/GENZ/... My assumption is that we will
rely on "vendor" support. But for what? The entire stack? Hooking into some API/Spec?

IMHO, if OFA is to survive, it needs to learn to be agile and know exactly what its offerings
are to be, and they should be very focused.

Steve...

On 8/31/17 21:47, Jim Ryan wrote:

Jason, because of my idiot gmail  system, I have to copy/paste your
comments below. I apologize for the redundancy this will likely present:


IBTA. We view interop as a program of value to participants who make the


biz


decision to fund it.


actual components are, AFAIK, left behind after testing. I have requested
another call for donations but, for whatever reason, that hasn't happened.


Ignoring donations, a logo program that pretty much exclusively
certifies discontinued equipment has deeply malfunctioned.



[JR] if that were the case, I'd have to agree, and I hope it's not. My


best info is there are new devices coming in for testing. If that's not the
case, I have no defense or further arguments.



I *do* have to ask you to not use terms along the lines of "membership
funding"; there is no such thing.


When I use that term, I am refering to the direct funding from the
membership of the OFA to the OFA treasury in the form of general dues,
IWG particpation fees, special contributions, and sponsorship
opportunities. There certainly is such a thing :)



[JR] I'm sorry, but I continue to disagree. OFA members make a specific


decision to pay dues and they make a separate decision as to whether to pay
logo/interop participation fees. I'm asking you to respect that and I'm
perplexed as to why you want to remove that option from them



Re the quality of testing, that's a challenge for the IWG. One of, if not


*the*


most important thing they're responsible for is quality of testing. If
something is broken there, I'm not aware of it, and we need to come to
understand this.


As far as granting logos to the submitted devices, it could be
fine.

As far as testing the OFED software and the open source stack around
it, it is vastly inadequate. I can say that confidently just from the
list of hardware being tested: It simply does not cover a very useful
(to end users) portion of the stack any longer.

So, again, I would like to see the OFA refocus this funding on better
testing. Scrap the logo program and ask the participating membership
to redirect the funding to direct software stack testing. Test the
software stack. Figure out how to directly buy modern hardware if
donations are not forthcoming. I hope this is the shape of the
discussion that is ongoing with the distros.

IMHO, this is how to get end-user orgs like RH, suse, LANL, etc to
particpate financially in the testing process.



[JR] I'm asking you to not lose this train of thought. I think this


could be really useful, w/o prejudging the outcome of the discussion. Plz
continue



Finally, I realized I failed to respond to a point you made earlier. It's


kinda


delicate, but important. The OFA is specifically not "chartered" to


develop


specs and the IBTA and others are. There are IP provisions that need to


exist


if this is part of our mission or not. I can give you boring details if


you


want to hear more.


Yes, as I said, I've argued this semantic point with Paul before ...



[JR] yes, again, we've walked this thin line. We can certainly address


this as part of our Bylaws review and that, of course, is the right way to
address this issue. Please don't give up; this could be extremely important
to us

Jim

I think this is something to fix in the new bylaws..



The reason this is delicate is because the OFIWG has had to go right
to the edge of what we can do, to use MAN pages to document expected
API functionality. We have agreed this is short of a spec, but you
get the point; it's a fine but important distinction.


.. because IP protections exist for a reason. Just because you call
the spec a MAN page, doesn't remove the need to be careful of IP
issues :)

Safeguarding against IP issues is an important role for an open source
foundation.
x

On Thu, Aug 31, 2017 at 7:13 PM, Jason Gunthorpe <
jgunthorpe at obsidianresearch.com<mailto:jgunthorpe at obsidianresearch.com>> wrote:



I broke the thread and revised the subject so it is easier to
follow. Thanks for suggesting it Paul.

On Thu, Aug 31, 2017 at 05:28:04PM -0700, Jim Ryan wrote:


Ok, I think I see your point. Your consciously blending membership dues


with


interop program participation to make a point. I don't want to argue that
point, but I do want to be painfully clear about something. The approach


we


take is conscious and, for example, specifically contrary to, for


example the

Yes, I at least, have always understood this is how IWG operates.



IBTA. We view interop as a program of value to participants who make the


biz


decision to fund it.


actual components are, AFAIK, left behind after testing. I have requested
another call for donations but, for whatever reason, that hasn't


happened.

Ignoring donations, a logo program that pretty much exclusively
certifies discontinued equipment has deeply malfunctioned.



I *do* have to ask you to not use terms along the lines of "membership
funding"; there is no such thing.


When I use that term, I am refering to the direct funding from the
membership of the OFA to the OFA treasury in the form of general dues,
IWG particpation fees, special contributions, and sponsorship
opportunities. There certainly is such a thing :)



Re the quality of testing, that's a challenge for the IWG. One of, if


not *the*


most important thing they're responsible for is quality of testing. If
something is broken there, I'm not aware of it, and we need to come to
understand this.


As far as granting logos to the submitted devices, it could be
fine.

As far as testing the OFED software and the open source stack around
it, it is vastly inadequate. I can say that confidently just from the
list of hardware being tested: It simply does not cover a very useful
(to end users) portion of the stack any longer.

So, again, I would like to see the OFA refocus this funding on better
testing. Scrap the logo program and ask the participating membership
to redirect the funding to direct software stack testing. Test the
software stack. Figure out how to directly buy modern hardware if
donations are not forthcoming. I hope this is the shape of the
discussion that is ongoing with the distros.

IMHO, this is how to get end-user orgs like RH, suse, LANL, etc to
particpate financially in the testing process.



Finally, I realized I failed to respond to a point you made earlier.


It's kinda


delicate, but important. The OFA is specifically not "chartered" to


develop


specs and the IBTA and others are. There are IP provisions that need to


exist


if this is part of our mission or not. I can give you boring details if


you


want to hear more.


Yes, as I said, I've argued this semantic point with Paul before ...

I think this is something to fix in the new bylaws..



The reason this is delicate is because the OFIWG has had to go right
to the edge of what we can do, to use MAN pages to document expected
API functionality. We have agreed this is short of a spec, but you
get the point; it's a fine but important distinction.


.. because IP protections exist for a reason. Just because you call
the spec a MAN page, doesn't remove the need to be careful of IP
issues :)

Safeguarding against IP issues is an important role for an open source
foundation.

Jason





_______________________________________________
Ofa_boardplus mailing list
Ofa_boardplus at lists.openfabrics.org<mailto:Ofa_boardplus at lists.openfabrics.org>
http://lists.openfabrics.org/mailman/listinfo/ofa_boardplus



--
Regards,
Steve...

May you be able to pursue what you love, yet excel at what you
must do.

_______________________________________________
Ofa_boardplus mailing list
Ofa_boardplus at lists.openfabrics.org<mailto:Ofa_boardplus at lists.openfabrics.org>
http://lists.openfabrics.org/mailman/listinfo/ofa_boardplus

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openfabrics.org/pipermail/ofa_boardplus/attachments/20170901/3fe51f0b/attachment.html>


More information about the Ofa_boardplus mailing list