Re: [Ospf-manet] MPR-OSPF GTNetS simulations report

All,

I ran simulations for 80 nodes, to add to my previous
GTNetS simulation results for 20, 40, and 60 nodes.
I used the same code that I used before.
INRIA's protocol used MPR-based LSA reduction and MPR-based
adjacency reduction (but retransmitted LSAs were not multicast).
MDR used 1-connected adjacencies and min-cost LSAs.
The same trend continued, with INRIA generating 15 times as much
DD overhead as MDR, and 5 times as much total overhead as MDR.
LS Ack overhead (not shown below) was 204.24 kbps for INRIA
and 75.14 kbps for MDR.
Note that using multicast retransmitted LSAs will not affect
DD overhead, and may increase LS Ack overhead (since every
adjacent neighbor must Ack each multicast LSA).
Parameters were given with my previous results.

Summary of results:

MDR                       20 nodes  40 nodes  60 nodes  80 nodes
---                       --------  --------  --------  --------
Total Overhead (kbps)      21.52     79.08     164.32   271.00
LSU overhead (kbps)         5.21     18.05      33.67    50.56
DD overhead (kbps)          4.08     18.54      36.28    60.88
Adjacencies/node            3.30      3.02      2.88      2.84
Adj changes/node/sec        0.04      0.04      0.04      0.04
Delivery ratio              0.966     0.951     0.953    0.949

INRIA                     20 nodes  40 nodes  60 nodes  80 nodes
-----                     --------  --------  --------  --------
Total Overhead (kbps)      48.20    234.12     588.43  1357.45 (5xMDR)
LSU overhead (kbps)        10.32     40.08      72.33   127.56
DD overhead (kbps)         18.40    119.92     352.89   934.31 (15xMDR)
Adjacencies/node            8.58     15.93      21.93    25.12
Adj changes/node/sec        0.15      0.27       0.36     0.53 (13xMDR)
Delivery ratio              0.918     0.924      0.920   0.912

Richard

_______________________________________________
Ospf-manet mailing list
Ospf-manet@ietf.org
https://www1.ietf.org/mailman/listinfo/ospf-manet