Roma Tre University
Is It Really Worth to
Peer at IXPs?
Marco Di Bartolomeo Giuseppe Di Battista
Roberto di Lallo Claudio Squarcella
UNIVERSITÀ DEGLI STUDI
ROMA TRE
Introduction to IXPs
Internet eXchange Points (IXPs) are infrastructures used by Internet Service Providers (ISPs) to
UNIVERSITÀ DEGLI STUDI ROMA TRE
Known Benefits
• Lower distances and locality of trafficKnown Benefits
• Lower distances
and
locality of traffic
UNIVERSITÀ DEGLI STUDI
ROMA TRE
Related Work (Selection)
• N. Chatzis, G. Smaragdakis, A. Feldmann, and W. Willinger,
“There is more to ixps than meets the eye”, SIGCOMM
Comput. Commun. 2013
• A. Gupta, L. Vanbever, M. Shahbaz, S. P. Donovan, B.
Schlinker, N. Feamster, J. Rexford, S. Shenker, R. Clark, and E. Katz-Bassett, “Sdx: A software defined internet
exchange”,in Proc. SIGCOMM 2014
• M. Ahmad and R. Guha,“A tale of nine internet exchange
points: Studying path latencies through major regional ixps”, in Local Computer Networks (LCN), 2012
UNIVERSITÀ DEGLI STUDI
ROMA TRE
Motivation and Contribution
MOTIVATION
• A quantitative analysis on the impact of IXPs on key performance
indicators is missing
• To our knowledge, all existing works use, at least partially, simulated data • Comparing a path traversing an IXP with one that does not is hard without
Motivation and Contribution
MOTIVATION
• A quantitative analysis on the impact of IXPs on key performance
indicators is missing
• To our knowledge, all existing works use, at least partially, simulated data • Comparing a path traversing an IXP with one that does not is hard without
direct control of traffic flows
OUR CONTRIBUTION
UNIVERSITÀ DEGLI STUDI
ROMA TRE
Side Motivation: Depeering
Some major ISPs have canceled all peerings (depeering) at the IXPs, justifying such decision in terms of :
• Cost reduction
UNIVERSITÀ DEGLI STUDI
ROMA TRE
QoS Investigation
Investigate the impact that IXPs have on the QoS
UNIVERSITÀ DEGLI STUDI
ROMA TRE
QoS Investigation
Round Trip Time
Investigate the impact that IXPs have on the QoS
USED METRICS
UNIVERSITÀ DEGLI STUDI
ROMA TRE
Hop Count
QoS Investigation
Round Trip Time
Investigate the impact that IXPs have on the QoS
UNIVERSITÀ DEGLI STUDI
ROMA TRE
Hop Count
QoS Investigation
Round Trip Time
Packet Loss
Investigate the impact that IXPs have on the QoS
USED METRICS
UNIVERSITÀ DEGLI STUDI
ROMA TRE
Jitter
Hop Count
QoS Investigation
Round Trip Time
Packet Loss
Investigate the impact that IXPs have on the QoS
UNIVERSITÀ DEGLI STUDI
ROMA TRE
Traffic Locality Investigation
Investigate the impact that IXPs have on keeping local the
A Case of Study
UNIVERSITÀ DEGLI STUDI
ROMA TRE
A Case of Study
A Case of Study
UNIVERSITÀ DEGLI STUDI
ROMA TRE
A Case of Study
Measuring System
We exploited the distributed infrastructure of RIPE Atlas, an Internet measurement network based on thousands of devices (probes), deployed all around the world.
UNIVERSITÀ DEGLI STUDI
ROMA TRE
Experiment CIS
Critical and most visited Internet Sites
Experiment CIS
Critical and most visited Internet Sites Most visited Internet Sites
• We selected the sites that are
most visited by italians
according to the Alexa ranking
• We filtered the ones hosted in
Italy
• Finally we selected the first
UNIVERSITÀ DEGLI STUDI
ROMA TRE
Experiment CIS: Setting
• Targets:
• 50 critical Internet services • 100 most visited sites
• Duration of the experiment: 4 hours • Frequency of measurements:
• 6 pings per hour
Experiment CIS: Results
Measurement: Round Trip Delay Targets: Critical Internet sites
0 % 20 % 40 % 60 % 80 % 100 % 0 20 40 60 80 100 120 140 160
Cumulated fraction of probes Probes not traversing IXPs
UNIVERSITÀ DEGLI STUDI ROMA TRE 0 % 20 % 40 % 60 % 80 % 100 % 6 8 10 12 14 16 18 20 22 24
Cumulated fraction of probes
Hop count
Probes not traversing IXPs probes traversing IXPs
Experiment CIS: Results
Measurement: Hop count
0 20 40 60 80 100 120
AS174AS3356AS8928AS3257AS1200AS2914AS1299AS6453AS8220AS3549AS1273AS20940AS34419
% of traversing paths
UPSTREAM IXP
Experiment CIS: Results
Measurement: Paths traversing foreign (non-Italian) ASes
UNIVERSITÀ DEGLI STUDI ROMA TRE 0 20 40 60 80 100 120
AS174 AS3356 AS2914 AS1299 AS3549 AS6939
% of traversing paths
UPSTREAM IXP
Experiment CIS: Results
Measurement: Paths traversing foreign (non-European) ASes
Experiment CIS: Discussion
• IXPs have a positive effect on keeping local the local
traffic
• As a first observation, paths through IXPs seem to have
better performance
• …However these measurements are performed by a
variable set of probes, since we don’t have direct
control on interdomain routing (i.e., BGP) announcements
UNIVERSITÀ DEGLI STUDI
ROMA TRE
Experiment SBA: Setting
Selective BGP Announcements UPSTREAM UPSTREAM PARTNER ISP PHASES: UPSTREAM IXP ALL Probe
Experiment SBA: Setting
Selective BGP Announcements UPSTREAM UPSTREAM PARTNER ISP PHASES: UPSTREAM IXP ALL ProbeUNIVERSITÀ DEGLI STUDI
ROMA TRE
Experiment SBA: Setting
Selective BGP Announcements UPSTREAM UPSTREAM PARTNER ISP PHASES: UPSTREAM IXP ALL Probe
Experiment SBA: Setting
Selective BGP Announcements UPSTREAM UPSTREAM PARTNER ISP PHASES: UPSTREAM IXP ALL ProbeUNIVERSITÀ DEGLI STUDI
ROMA TRE
Experiment SBA: Setting
Selective BGP Announcements UPSTREAM UPSTREAM PARTNER ISP PHASES: UPSTREAM IXP ALL Probe
Experiment SBA: Setting
Selective BGP Announcements UPSTREAM UPSTREAM PARTNER ISP PHASES: UPSTREAM IXP ALL ProbeUNIVERSITÀ DEGLI STUDI
ROMA TRE
Experiment SBA: Setting
Selective BGP Announcements UPSTREAM UPSTREAM PARTNER ISP PHASES: UPSTREAM IXP ALL Probe
UPSTREAM UPSTREAM PARTNER ISP PHASES: UPSTREAM IXP ALL Probe
Experiment SBA: Setting
UNIVERSITÀ DEGLI STUDI
ROMA TRE
Partner ISPs
For the second experiment we partnered with 3 providers:
Mc-Link
Italian network operator headquartered in Rome
and founded in 1986. It mainly offers services
for the business industry.
Seeweb
Italian network operator headquartered in Rome
and founded in 1998. It offers cloud services
and web hosting
Unidata
Italian network operator headquartered in Rome
and founded in 1985. Among the co-founders
of the popular IXP NaMeX
0 % 20 % 40 % 60 % 80 % 100 % 0 10 20 30 40 50 60 70 80
Cumulated fraction of probes
Probes choosing upstream (M) Probes choosing IXPs (M) Probes choosing upstream (F) Probes choosing IXPs (F)
Experiment SBA: Results
Measurement: Average Round Trip Delay Target(s): Inside Seeweb (“M” and “F”)
UNIVERSITÀ DEGLI STUDI ROMA TRE 0 % 20 % 40 % 60 % 80 % 100 % 0 10 20 30 40 50 60 70
Cumulated fraction of probes
Average round-trip delay [ms] AS 174
IXP
Experiment SBA: Results
Measurement: Average Round Trip Delay
0 % 20 % 40 % 60 % 80 % 100 % 2 4 6 8 10 12 14 16 18 20
Cumulated fraction of probes Probes choosing upstream
Probes choosing IXPs
Experiment SBA: Results
Measurement: Average Hop count Target: Inside MC-LINK
UNIVERSITÀ DEGLI STUDI ROMA TRE 0 % 20 % 40 % 60 % 80 % 100 % 5 6 7 8 9 10 11 12 13
Cumulated fraction of probes
Average hop count AS 3356
IXP
Experiment SBA: Results
Measurement: Average Hop count
0 % 20 % 40 % 60 % 80 % 100 % 0 50 100 150 200 250
Cumulated fraction of probes Probes choosing upstream
Probes choosing IXPs
Experiment SBA: Results
Measurement: Jitter Target: Inside Unidata
UNIVERSITÀ DEGLI STUDI ROMA TRE 0 % 20 % 40 % 60 % 80 % 100 % 0 50 100 150 200 250
Cumulated fraction of probes
Jitter [ms]
AS 3257 IXP
Experiment SBA: Results
Measurement: Jitter
Experiment SBA: Results
Measurement: Gain on the RTT (per class of probes) Target: Inside MC-LINK
0 % 20 % 40 % 60 % 80 % 100 % -60 -50 -40 -30 -20 -10 0 10
Cumulated fraction of probes
CAH Customer Transit
UNIVERSITÀ DEGLI STUDI
ROMA TRE
Experiment SBA: Results
Measurement: Gain on the Hop Count (per class of probes) Target: Inside MC-LINK
0 % 20 % 40 % 60 % 80 % 100 % -10 -9 -8 -7 -6 -5 -4 -3 -2 -1
Cumulated fraction of probes
Gap in average hop count CAH
Customer Transit
Conclusions
Our experiments put in evidence that peerings exploiting
IXPs:
• have a positive effect on key performance indicators
• have the effect of reducing the number of foreign ISPs
traversed to reach critical Internet services (keep local
UNIVERSITÀ DEGLI STUDI
ROMA TRE
Future Work
It would be interesting to:
• extend our experiments considering more performance
indicators (e.g. bandwidth)
• cooperate with ISPs and IXPs in other countries to
reproduce the experiments, compare the results and