[codes-ross-users] how to use fattree static routing
Nikhil Jain
nikhil.jain at acm.org
Mon Mar 13 16:39:40 CDT 2017
routing_folder and dot_file are keywords/parameters in the config file whose values were chosen by you during the topology dump run. It is described in the second step "2- Static Routing”.
---
Nikhil Jain
Postdoctoral Fellow, Lawrence Livermore National Laboratory
nikhil.jain at acm.org, http://nikhil-jain.github.io/
> On Mar 13, 2017, at 14:25, Peixin Qiao <pqiao at hawk.iit.edu> wrote:
>
> When I run the third step as README.fattree.txt:
>
> # 3. Choose a routing algorithm which should be used by OpenSM
> # (possible options: updn, dnup, ftree, lash, dor, torus-2QoS, dfsssp, sssp)
> # export OSM_ROUTING="ftree"
> # ~/simulation/scripts/create_static_lft.sh routing_folder dot_file
>
> # (here routing_folder and dot_file should be same as the one used during the run used to dump the topology)
>
> # Now, the routing table stored as LFT files should be in the routing_folder.
>
> So what is the routing_folder and dot_file? Is the following correct?
>
> routing_folder: mkdir routing_folder
> dot_file: python simulate.py createIBNet -t XGFT -xh 3 -xm 18,18,36 -xw 1,18,18 -sp 36 (get 2 files: net.dot and net.txt)
>
> Thanks,
> Peixin
>
> On Mon, Mar 13, 2017 at 1:52 PM, Nikhil Jain <nikhil.jain at acm.org <mailto:nikhil.jain at acm.org>> wrote:
> Did the script die or you killed it? If you killed it, after how long? What size system are you generating routing for?
>
> ---
> Nikhil Jain
> Postdoctoral Fellow, Lawrence Livermore National Laboratory
> nikhil.jain at acm.org <mailto:nikhil.jain at acm.org>, http://nikhil-jain.github.io/ <http://nikhil-jain.github.io/>
>> On Mar 13, 2017, at 10:31, Peixin Qiao <pqiao at hawk.iit.edu <mailto:pqiao at hawk.iit.edu>> wrote:
>>
>> Hi CODES,
>>
>> When I run the command in README.fattree.txt, I cannot finish creating static routing table. The third step cannot be finished as follows:
>>
>> peixin at peixin:~/simulation/scripts/routing_folder$ ls
>> net.dot net.txt
>> peixin at peixin:~/simulation/scripts/routing_folder$ cd ..
>> peixin at peixin:~/simulation/scripts$ ~/simulation/scripts/create_static_lft.sh /home/peixin/simulation/scripts/routing_folder/ net
>> dot file already exists.
>> running createIBNet.py
>> Selected configuration:
>> Output: /home/peixin/simulation/scripts/routing_folder/topo.net <http://topo.net/>
>> Topology: XGFT
>> Number of endpoints: 8
>> Number of ports per endpoints: 1
>> Number of switches: 4
>> Number of ports per switch: 36
>> Dimensions for Mesh/Torus: -1, -1, -1
>> k-ary-n Tree: -1 -ary- -1
>> Kautz K(b,n): K(-1; -1)
>> XGFT(h,m,w): XGFT(3; 18,18,36; 1,18,18)
>> Dragonfly(a,p,h) : Dragonfly(-1, -1, -1) with -1 groups
>> Cascade(a,p,h,g) : Cascade(96, 8, 10) with -1 groups
>> Tofu(d1,d2,d3) : Tofu(-1, -1, -1) with -1 groups
>>
>> This program also writes an dot-file to plot the graph with Graphviz.
>> Try:
>> dot / neato /home/peixin/simulation/scripts/routing_folder/topo.dot -Tpng -o graph.png && evince graph.png
>> running simulate.py
>> Run ibsim...
>> Run opensm...
>> ^CTraceback (most recent call last):
>> File "/home/peixin/simulation/scripts/simulate.py", line 1140, in <module>
>> app = runSim(options.network, options.routing, options.pattern)
>> File "/home/peixin/simulation/scripts/simulate.py", line 826, in __init__
>> time.sleep(60)
>> KeyboardInterrupt
>> peixin at peixin:~/simulation/scripts$ cd routing_folder/
>> peixin at peixin:~/simulation/scripts/routing_folder$ ls
>> log.txt net.dot net.txt ofedout rootguids.conf topo.dot topo.net <http://topo.net/>
>> peixin at peixin:~/simulation/scripts/routing_folder$
>>
>>
>>
>> How to create fat-tree static routing table with 36-ports switch?
>>
>> Thanks,
>> Peixin
>>
>> _______________________________________________
>> codes-ross-users mailing list
>> codes-ross-users at lists.mcs.anl.gov <mailto:codes-ross-users at lists.mcs.anl.gov>
>> https://lists.mcs.anl.gov/mailman/listinfo/codes-ross-users <https://lists.mcs.anl.gov/mailman/listinfo/codes-ross-users>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/codes-ross-users/attachments/20170313/17d7e80b/attachment-0001.html>
More information about the codes-ross-users
mailing list