Re: FabricPath in DCN

Thanks guys for the response.
Packetpushers podcast and forum is a good resource for these emerging tech.

Regards,
Alexander Lim

On 28 Feb, 2013, at 2:41 AM, marc edwards <renorider_at_gmail.com> wrote:

> I think to answer the original question of technologies such as
> TRILL/Fabric Path/ newly learned SPB will become legacy with
> introduction of SDN would be it depends. It all depends on timing. If
> the DC is large enough and a nexus solution has been rolled out, it
> might make sense to do things now instead of wait until SDN matures to
> point where almost everything runs on flow based architecture.
>
> All three seem to take the same idea of using L3 routing protocol in
> L2 space to provide load share over links. Also more dynamic when
> compared to traditional STP/MST.
>
> My DC isn't large enough to have this concern. We have been moving
> towards Nexus but didn't buy into FabricPath licencsing. It was too
> expensive for what we would use it for. On that note, (if SE is on
> list) why would I even consider Fabric Path license if I could instead
> go with open standard TRILL. I can only assume that both features are
> unlocked under the same lic.
>
> Marc Abel referenced packet pushers site for learning about SDN. I
> have put more attention to this site. Ethan Banks has become a regular
> name in daily activities. Fact is, if we aren't keeping up to speed on
> SDN we WILL be obsolete. Seems to be the next evolution in networking
> and where my focus will be.
>
> Great discussion. I know it is off topic for R&S for blueprint now but
> something I see every CCIE R&S having to visit in one form or another
> during recertification.
>
> Regards,
>
> Marc
>
> On Wed, Feb 27, 2013 at 9:37 AM, Murphy, William
> <William.Murphy_at_uth.tmc.edu> wrote:
>> We recently deployed fabricpath in our data center and it is working well so far, although we have not migrated all of our servers over to Nexus yet... We selected fabicpath to provide super-fast layer-2 spine within the room to provide lots of bandwidth rack-to-rack and row-to-row within the data center... We have no plans to use it for DCI or other applications right now... It is easier to manage VLANs and I would say it is now easier to turn-up services VLANs because there is no trunking/pruning of VLANs required as we had to do in the old Catalyst environment... We have tested things like VMotion and there is a substantial improvement on the order of 50s reduced to 10s to move a workload... Some of this is obviously attributed to the switch to 10G, but as we migrate all our services to Nexus we expect to maintain this kind of performance... Happy customers so far...
>>
>> Bill
>>
>> -----Original Message-----
>> From: nobody_at_groupstudy.com [mailto:nobody_at_groupstudy.com] On Behalf Of Alexander Lim
>> Sent: Tuesday, February 26, 2013 5:27 PM
>> To: Cisco certification
>> Subject: OOT: FabricPath in DCN
>>
>> Guys,
>>
>> m(_ _)m Sorry it's OOT. Please let me know if there is a more appropriate forum/mailing-list for this kind of discussion.
>>
>> I am interested if any of you have implemented FabricPath in your data centre network? Or other similar technology: TRILL/SPB 'compatible' from other vendor?
>> Do you think it is the way to go for DCN? Or it is just a transit technology to other tech, i.e. SDN?
>> Thank you for sharing.
>>
>> Regards,
>> Alexander Lim
>>
>>
>> Blogs and organic groups at http://www.ccie.net
>>
>> _______________________________________________________________________
>> Subscription information may be found at:
>> http://www.groupstudy.com/list/CCIELab.html
>>
>>
>> Blogs and organic groups at http://www.ccie.net
>>
>> _______________________________________________________________________
>> Subscription information may be found at:
>> http://www.groupstudy.com/list/CCIELab.html

Blogs and organic groups at http://www.ccie.net
Received on Mon Mar 04 2013 - 08:28:42 ART