[ofa-general] Re: [PATCH OpenSM 0/3] Fat Tree - Routing between non-CN nodes

[Yevgeny Kliteynik]

Hal Rosenstock wrote:
> On Wed, Feb 11, 2009 at 8:26 AM, Yevgeny Kliteynik
> <kliteyn at dev.mellanox.co.il> wrote:
>> Sasha Khapyorsky wrote:
>>> On 10:25 Wed 11 Feb     , Nicolas Morey Chaisemartin wrote:
>>>> What about high nodes (HN) as it concerns only nodes which are not at the
>>>> bottom of the fat tree?
>>> Could be fine. Let's ask Yevgeny too... :)
>>>
>>> Yevgeny! Any idea about io_nodes more generic name?
>> Ugh...
>>
>> "IO nodes":
>> Pros: the name is closer to the reality, since in most cases
>> the nodes that would need special treatment are indeed IO nodes.
>> Cons: the name is not "general"...
>>
>> "High nodes"
>> Pros: general name with kinda "hint" to the special treatment.
>> Cons: the "hint" is rather vague...
>>
>> Bottom line - I'm OK with both options (slightly leaning toward IO),
>> as long as it is described well enough in the help message and in man :)
> 
> Maybe consistency is the hobgobblin of small minds but don't we now have:
> 
> high nodes which is a topology based name
> and
> compute nodes which is a functional based name.
> 
> Is it worth having them consistent ?

Good point. IO nodes will be consistent with CNs.

-- Yevgeny

> -- Hal
> 
>> -- Yevgeny
>>
>>> Sasha
>>>
>> _______________________________________________
>> general mailing list
>> general at lists.openfabrics.org
>> http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
>>
>> To unsubscribe, please visit
>> http://openib.org/mailman/listinfo/openib-general
>>
>