[Mini-NDN] topology result
Lan Wang (lanwang)
lanwang at memphis.edu
Fri Nov 25 13:01:50 PST 2016
Also the figure from the paper shows a round trip time of around 5ms so you should set your link delay accordingly to get this RTT (not 20ms as suggested earlier).
Lan
On Nov 24, 2016, at 2:38 PM, Ashlesh Gawande (agawande) <agawande at memphis.edu<mailto:agawande at memphis.edu>> wrote:
Yes run the experiment with different faces. No need to write anything.
Ashlesh
________________________________
From: sepehr abdous <sepehrabdous1375 at gmail.com<mailto:sepehrabdous1375 at gmail.com>>
Sent: Thursday, November 24, 2016 2:27:18 PM
To: Ashlesh Gawande (agawande)
Cc: mini-ndn at lists.cs.ucla.edu<mailto:mini-ndn at lists.cs.ucla.edu>
Subject: Re: [Mini-NDN] topology result
about (2) you mean that I should write the following experiment with different faces?
from ndn.experiments.experiment import Experiment
from ndn.nlsr import Nlsr
import time
class fail(Experiment):
def __init__(self, args):
args["nPings"] = 140000
Experiment.__init__(self, args)
self.PING_COLLECTION_TIME_BEFORE_FAILURE = 60
self.PING_COLLECTION_TIME_AFTER_RECOVERY = 20
def run(self):
self.startPctPings()
# After the pings are scheduled, collect pings for 1 minute
time.sleep(self.PING_COLLECTION_TIME_BEFORE_FAILURE)
# Bring down CSU
for host in self.net.hosts:
if host.name<http://host.name/> == "node4":
self.failNode(host)
break
# CSU is down for 2 minutes
time.sleep(120)
# Bring CSU back up
for host in self.net.hosts:
if host.name<http://host.name/> == "node4":
self.recoverNode(host)
for other in self.net.hosts:
if host.name<http://host.name/> != other.name<http://other.name/>:
self.ping(host, other, self.PING_COLLECTION_TIME_AFTER_RECOVERY)
# Collect pings for more seconds after CSU is up
time.sleep(self.PING_COLLECTION_TIME_AFTER_RECOVERY)
Experiment.register("failone", fail)
[https://ssl.gstatic.com/ui/v1/icons/mail/images/cleardot.gif]
On Thu, Nov 24, 2016 at 10:32 AM, Ashlesh Gawande (agawande) <agawande at memphis.edu<mailto:agawande at memphis.edu>> wrote:
1) I think the experiment from paper was run on real testbed which would explain the outliers in the RTT.
2) The experiment for blue line is not correct as it also a failure experiment. The difference is the number of faces.
For blue line the experiment would be the same as the red line except you would run mini-NDN with option --faces 0 (i.e. multipath). For the red line you would run with --faces 1 (i.e. single path).
Blue line:
sudo minindn --experiment=failureExp --faces 0
Red line:
sudo minindn --experiment=failureExp --faces 1
3) I am not sure about the density but the figure on the paper should not have more than 200 pings. Maybe the small size of the figure and different point size is leading to more observed density.
Another case could be that there are some timeouts in your experiment that are not represented on the graphs?
Ashlesh
________________________________
From: Mini-NDN <mini-ndn-bounces at lists.cs.ucla.edu<mailto:mini-ndn-bounces at lists.cs.ucla.edu>> on behalf of sepehr abdous <sepehrabdous1375 at gmail.com<mailto:sepehrabdous1375 at gmail.com>>
Sent: Wednesday, November 23, 2016 4:06:01 PM
To: mini-ndn at lists.cs.ucla.edu<mailto:mini-ndn at lists.cs.ucla.edu>
Subject: Re: [Mini-NDN] topology result
also my experiments codes are:
for blue points:
from ndn.experiments.experiment import Experiment
from ndn.nlsr import Nlsr
import time
class pingallreg(Experiment):
def __init__(self, args):
args["nPings"] = 200000
Experiment.__init__(self, args)
def run(self):
self.startPctPings()
time.sleep(200)
Experiment.register("wofail", pingallreg)
and for red points which involve failure:
from ndn.experiments.experiment import Experiment
from ndn.nlsr import Nlsr
import time
class fail(Experiment):
def __init__(self, args):
args["nPings"] = 140000
Experiment.__init__(self, args)
self.PING_COLLECTION_TIME_BEFORE_FAILURE = 60
self.PING_COLLECTION_TIME_AFTER_RECOVERY = 20
def run(self):
self.startPctPings()
# After the pings are scheduled, collect pings for 1 minute
time.sleep(self.PING_COLLECTION_TIME_BEFORE_FAILURE)
# Bring down CSU
for host in self.net.hosts:
if host.name<http://host.name/> == "node4":
self.failNode(host)
break
# CSU is down for 2 minutes
time.sleep(120)
# Bring CSU back up
for host in self.net.hosts:
if host.name<http://host.name/> == "node4":
self.recoverNode(host)
for other in self.net.hosts:
if host.name<http://host.name/> != other.name<http://other.name/>:
self.ping(host, other, self.PING_COLLECTION_TIME_AFTER_RECOVERY)
# Collect pings for more seconds after CSU is up
time.sleep(self.PING_COLLECTION_TIME_AFTER_RECOVERY)
Experiment.register("failone", fail)
On Thu, Nov 24, 2016 at 1:31 AM, sepehr abdous <sepehrabdous1375 at gmail.com<mailto:sepehrabdous1375 at gmail.com>> wrote:
Hello all,
I'm new to mini-ndn and I'm trying to implement the following topology
<Untitled.png>
I implemented it with the following information:
<Untitled(1).png>
and based on the paper the expected results on ping is :
<Untitled(2).png>
but when I implement it I get the following image
<image(1).png>
and as you can see there are some problems:
(1) the values on the y-axis is much higher than in the paper, I figured out that It's caused by the delay, I set the delay for every link to 20ms so the values are around 80ms but when I set the delay to 0ms the values are around 10ms but the swing of values become more and values are not united like when the delay is 20ms
(2) as you can see when node4 fails we have some values that shows something happened:
<NLSR.jpg>
but this values doesn't exist in my figure.
(3) the density of red points in my figure isn't the same in all aspects.
can you please help me with this problems.
Thank you
--
Sepehr Abdous
Faculty: abdous at ce.sharif.edu<mailto:abdous at ce.sharif.edu>
Yahoo mail: sepehrabdous at yahoo.com<mailto:sepehrabdous at yahoo.com>
Cellphone:+989194030591
School of computer engineering
Sharif University of Technology
Tehran, Iran
--
Sepehr Abdous
Faculty: abdous at ce.sharif.edu<mailto:abdous at ce.sharif.edu>
Yahoo mail: sepehrabdous at yahoo.com<mailto:sepehrabdous at yahoo.com>
Cellphone:+989194030591
School of computer engineering
Sharif University of Technology
Tehran, Iran
_______________________________________________
Mini-NDN mailing list
Mini-NDN at lists.cs.ucla.edu<mailto:Mini-NDN at lists.cs.ucla.edu>
http://www.lists.cs.ucla.edu/mailman/listinfo/mini-ndn
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/mini-ndn/attachments/20161125/fde9c84f/attachment-0001.html>
More information about the Mini-NDN
mailing list