07-08-2015 03:21 PM
The Cisco Network Convergence System 6000 (NCS 6000) helps enable superior network agility, packet optical convergence, and petabits-per-second system scale. The series facilitates the Evolved Programmable Network to support virtualization and programmability at the lowest total cost of ownership, while delivering high-bandwidth mobile, video, and cloud services to end users.
Ask questions from July 20, 2015 through July 31, 2015
Featured Expert
Nikhil Kapoor is currently working at Cisco HTTS (High Touch Technical Support ) under the IOS-XR technologies. I have been supporting customers and partners with issues on ASR9k,CRS,CRS Multi Chassis, NCS. I currently hold CCIE Service provide CCIE #45281 and also hold Cisco IOS-XR Specialist. Within Cisco, I have been working with the HTTS IOS XR team providing various trainings, developing internal knowledge and articles to help the engineers on IOS and IOS-XR platforms.
We look forward to your participation. This event is open to all, including partners. Please Share this event in your social channels.
Find other https://supportforums.cisco.com/expert-corner/events.
**Ratings Encourage Participation! **
Please be sure to rate the Answers to Questions
Solved! Go to Solution.
07-28-2015 07:28 PM
Hi Ed,
Thanks for question. NCS6008 can have 6 fabric cards (PID NC6-FC-S) and 5 planes provide full capacity. Performs 2.1 Billion cells per second per fabric ASIC. Each fabric card will have 2 ASIC. The forwarding over fabric is via cells like we do in CRS. Each fabric (Ingress FIA) request the destination fabric (Egress FIA) to send/ pass the cells once the access is granted virtual output queues are established which pass the cells through fabric towards destination. At this stage we start chopping the packets in to cells and finally re-assembled at Egress FIA. Yes this is somewhat same as we have in CRS.
Thanks,
Nikhil
07-21-2015 08:26 AM
Hi Nikhil
We will be soon deploying NCS and have started working on it. Not sure If I am taking the correct word but could you let us know how NCS works on Green mode?
Thanks
07-21-2015 08:55 AM
Hello,
Thanks for the question. I completely understand your query with use of NCS power consumption can be done. Going to LC architecture we have 5 slices each of 200 Gbps which computes to 1Tbytes per LC. . In an event we are not using any interface we can shutdown the slice, consequently saving the power. This is known as Green Mode for LC NCS. Hope this answers your query.
Thanks,
Nikhil
07-21-2015 09:01 AM
Thanks Nikhil and Osman.
I may want to ask one more question. Is 200 Gbps uni-directional or Bi-directional?
07-21-2015 09:03 AM
Hello,
200Gbps per slice is bi-directional.
Thanks,
Nikhil
07-21-2015 08:56 AM
Hi
You may review some information on the green mode within the release notes, e.g. for 5.2.3:
http://www.cisco.com/c/en/us/td/docs/routers/ncs6000/software/release/notes/reln_523ncs.html
We can shut down a slice using a config CLI. The interfaces belonging to the slice that is shut down will not be visible. Entity MIB will show that the interfaces have been removed from the system.
CLI:
(config) # hw-module location <> slice <> shutdown
07-21-2015 09:26 AM
Hi Nikhil,
Does NCS supports multichassis? How about B2B
Thanks
Amit
07-21-2015 09:37 AM
Hi Amit,
Yes Multi Chassis is possible with NCS 6008 in a same fashion like we do for CRS via Control Ethernet connections.
Thanks,
Nikhil
07-21-2015 10:23 AM
Thanks Nikhil and Osman.
07-21-2015 10:01 AM
Please use the following document to get started:
http://www.cisco.com/c/en/us/td/docs/routers/ncs6000/fabricchassis/hig/guide/fabricHIG.html
07-21-2015 10:35 AM
Hi Again,
A quick one For Multichassis and B2B could let us know the optics supported?
Thanks,
Amit
07-21-2015 10:43 AM
Hi Amit,
Thanks for questions. Please feel free to ask as many questions. For Multi Chassis and B2B on NCS6008 can be achieved via CXP optical interconnects.
Hope this helps.
Thanks,
Nikhil
07-22-2015 08:46 AM
Hi Nikhil,
I have been going through few docs and came across question. If you could help. Is there any way that we can breakout 100G LC in 10x10G and how could we check the port to slice mapping?
Thanks,
07-22-2015 08:59 AM
Hello,
Thanks for question. Please feel free to ask as many questions. Coming to your query yes there is way to break out 1x100G in 10X10G.
Command to do so :
RP/0/RP0/CPU0:6008-A(config)#hw-module location 0/0/cpU0 slice 1 breakout 10G
For slice mapping we can execute below command.
RP/0/RP0/CPU0:6008-A#show uidb index location 0/0/CpU0
Example output of command for slice mapping.
<SNIP>
Location Interface-name Interface-Type NPU UIDB-indices
--------------------------------------------------------------------------------
0/0/CPU0 TenGigE0_0_0_0 Main Interface 0 1
0/0/CPU0 TenGigE0_0_0_1 Main Interface 0 2
0/0/CPU0 TenGigE0_0_0_2 Main Interface 0 3
0/0/CPU0 TenGigE0_0_0_3 Main Interface 0 4
Hope this helps.
Thanks,
Nikhil
07-22-2015 09:34 AM
Thanks Nikhil
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide