revised EPC profiling docs

This commit is contained in:
Nicola Baldo
2011-12-11 13:50:30 +01:00
parent 4ed8385514
commit 580f44e8fd
2 changed files with 4 additions and 4 deletions

View File

@@ -25,8 +25,7 @@ import sys, os
# Add any Sphinx extension module names here, as strings. They can be extensions
# coming with Sphinx (named 'sphinx.ext.*') or your custom ones.
extensions = ['sphinx.ext.pngmath',
'sphinxcontrib.seqdiag']
extensions = ['sphinx.ext.pngmath']
# Add any paths that contain templates here, relative to this directory.
templates_path = ['_templates']

View File

@@ -166,7 +166,7 @@ Running time evolution is quadratic since we increase at the same time the numbe
Running time
As a rough complexity estimation, we compare two scenarios using the simplified E-UTRAN version (PHY+MAC) against the complete E-UTRAN + EPC, with same number of UEs per eNodeB, same number of eNodeBs and roughly same volume of transmitted data.
To estimate the additional complexity of the upper LTE Radio Protocol Stack model and the EPC model, we compare two scenarios using the simplified E-UTRAN version (using only PHY, MAC and the simplified RLC/SM, with no EPC and no ns-3 applications) against the complete E-UTRAN + EPC (with UM RLC, PDCP, end-to-end IP networking and regular ns-3 applications). Both configuration have been tested with the same number of UEs per eNodeB, the same number of eNodeBs, and approximately the same volume of transmitted data (an exact match was not possible due to the different ways in which packets are generated in the two configurations).
.. _fig-epcEutranRunTime:
@@ -176,7 +176,8 @@ As a rough complexity estimation, we compare two scenarios using the simplified
EPC E-UTRAN running time
Figures for the complete system are more or less doubling the previous one, but considering the overhead of all the logic and entities incorporated this order of increment is to be expected.
From the figure, it is evident that the additional complexity of using the upper LTE stack plus the EPC model translates approximately into a doubling of the execution time of the simulations. We believe that, considered all the new features that have been added, this figure is acceptable.
Simulation time
---------------