1 cps 214 computer networks and distributed systems “live” video and audio streaming end system...
TRANSCRIPT
![Page 1: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/1.jpg)
1
CPS 214Computer Networks and Distributed Systems
“Live” Video and Audio Streaming•End System Multicast•Analysis of Akamai Workload
![Page 2: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/2.jpg)
2
Presentations• Monday, April 21
– Abhinav, Risi– Bi, Jie– Jason, Michael– Martin, Matt– Amre, Kareem
• Wednesday, April 23– Ben, Kyle– Jayan, Michael– Kshipra, Peng– Xuhan, Yang
![Page 3: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/3.jpg)
The Feasibility of Supporting Large-Scale Live Streaming Applications with Dynamic
Application End-Points
Kay Sripanidkulchai,
Aditya Ganjam, Bruce Maggs*, and Hui Zhang
Carnegie Mellon University
* and Akamai Technologies
![Page 4: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/4.jpg)
4
Motivation• Ubiquitous Internet broadcast
– Anyone can broadcast– Anyone can tune in
![Page 5: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/5.jpg)
5
Overlay multicast architectures
RouterSourceApplication end-point
![Page 6: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/6.jpg)
6
Infrastructure-based architecture[Akamai]
+ Well-provisionedRouterSourceApplication end-pointInfrastructure server
![Page 7: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/7.jpg)
7
Application end-point architecture[End System Multicast (ESM)]
+ Instantly deployable+ Enables ubiquitous broadcast
RouterSourceApplication end-point
![Page 8: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/8.jpg)
8
Waypoint architecture [ESM]
+ Waypoints as insurance RouterSourceApplication end-pointWaypoint
W
W
![Page 9: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/9.jpg)
9
Sample ESM Broadcasts http://esm.cs.cmu.edu
Event Duration(hours)
Unique Hosts
Peak Size
SIGCOMM ’02 25 338 83
SIGCOMM ’03 72 705 101
SOSP’03 24 401 56
DISC’03 16 30 20
Distinguished Lectures
11 400 80
AID Meeting 14 43 14
Buggy Race 24 85 44
Slashdot 24 1609 160
Grand Challenge
6 2005 280
![Page 10: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/10.jpg)
10
Feasibility of supporting large-scale groups with an application end-point architecture?
• Is the overlay stable enough despite dynamic participation?
• Is there enough upstream bandwidth?• Are overlay structures efficient?
![Page 11: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/11.jpg)
11
Large-scale groups
• Challenging to address these fundamental feasibility questions– Little knowledge of what large-scale live streaming
is like
![Page 12: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/12.jpg)
12
Chicken and egg problem
Publishers with compelling content need proof that the system works.
System has not attracted large-scale groups due to lack of compelling content.
![Page 13: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/13.jpg)
13
The focus of this paper
• Generate new insight on the feasibility of application end-point architectures for large scale broadcast
• Our methodology to break the cycle– Analysis and simulation– Leverage an extensive set of real-world workloads
from Akamai (infrastructure-based architecture)
![Page 14: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/14.jpg)
14
Talk outline
• Akamai live streaming workload• With an application end-point architecture
– Is the overlay stable enough despite dynamic participation?
– Is there enough upstream bandwidth?– Are overlay structures efficient?
• Summary
![Page 15: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/15.jpg)
15
Measurements used in this study
• Akamai live streaming traces– Trace format for a request
[IP, Stream URL, Session start time, Session duration]
• Additional measurements collected– Hosts’ upstream bandwidth
![Page 16: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/16.jpg)
An Analysis of Live Streaming Workloads on the Internet
Kunwadee Sripanidkulchai, Bruce Maggs*, Hui ZhangCarnegie Mellon University and
*Akamai Technologies
![Page 17: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/17.jpg)
17
Akamai live streaming infrastructure
A
A
A
A
A
A
Reflectors Edge servers
Source
![Page 18: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/18.jpg)
18
Extensive traces
~ 1,000,000 daily requests ~ 200,000 daily client IP addresses from over
200 countries~ 1,000 daily streams~ 1,000 edge servers~ Everyday, over a 3-month period
![Page 19: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/19.jpg)
19
Largest stream
75,000 x 250 kbps = 18 Gbps!
![Page 20: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/20.jpg)
20
Highlight of findings
• Popularity of events [Bimodal Zipf]
• Session arrivals [Exponential for short time-scales, time-of-day and time-zone-correlated behavior, LOTS of flash crowds]
• Session durations [Heavy-tailed]
• Transport protocol usage [TCP rivals UDP]
• Client lifetime• Client diversity
![Page 21: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/21.jpg)
21
Request volume (daily)
Weekdays
Weekends
Missing logs
Num
ber
of r
eque
sts
![Page 22: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/22.jpg)
22
Audio vs. video
Unknown 22%
Audio 71%
Video 7%
Most streams are audio.
![Page 23: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/23.jpg)
23
Stream types• Non-stop (76%) vs. short duration (24%)
– All video streams have short duration
• Smooth arrivals (50%) vs. flash crowds (50%)– Flash crowds are common
![Page 24: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/24.jpg)
24
Client lifetime
• Motivating questions– Should servers maintain “persistent” state about
clients (for content customization)?– Should clients maintain server history (for server
selection problems)?
• Want to know– Are new clients tuning in to an event?– What is the lifetime of a client?
![Page 25: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/25.jpg)
25
Analysis methodology
• Windows media format• Player ID field to identify distinct users
• Birth rate = Number of new distinct users
Total number of distinct users
![Page 26: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/26.jpg)
26
Daily new client birth rate
• New client birth rate is 10-100% across all events.• For these 2 events, birth rate is 10-30%.
Weekends
Weekdays Xmas
![Page 27: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/27.jpg)
27
One-timers: tune in for only 1 day
In almost all events, 50% of clients are one-timers!
![Page 28: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/28.jpg)
28
Client lifetime (excluding one-timers)
y = x
y = 3x
For most events, average client lifetime is at least 1/3 of the event duration.
![Page 29: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/29.jpg)
29
Client lifetime
• Motivating questions– Should servers maintain “persistent” state about
clients (for content customization)? Any state should time-out quickly because most clients are one-timers.
– Should clients maintain server history (for server selection problems)? Yes, recurring clients tend to hang around for a while.
![Page 30: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/30.jpg)
30
Where are clients from?
0.00E+00
2.00E+05
4.00E+05
6.00E+05
8.00E+05
1.00E+06
1.20E+06
1.40E+06
1.60E+06 US
CN
DE
ES
FR
GB
CA
JP
PT
CH
BE
MX
NL
SE
KR
BR
Countries
Num
ber
of I
P A
ddre
sses
Clients are from over 200 countries.Most clients are from the US and Europe.
![Page 31: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/31.jpg)
31
Analysis methodology
• Map client IP to location using Akamai’s EdgeScape tool
• Definitions– Diversity index = Number of distinct ‘locations’ that
a stream reaches– Large streams are streams that have a peak group
size of more than 1,000 clients
![Page 32: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/32.jpg)
32
Time zone diversity
Almost all large streams reach more than half the world.
Many small streams reach more than half the world!
![Page 33: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/33.jpg)
33
Client diversity
• Motivating questions– Where should streaming servers be placed in the
network? Clients are tuning in from many different locations.
– How should clients be mapped to servers? For small streams which happen to have a diverse set of clients, it may be too wasteful for a CDN to map every client to the nearest server.
![Page 34: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/34.jpg)
34
Summary
• Publishers are using the Internet to reach a wider audience than traditional radio and TV
• Interesting observations– Lots of audio traffic– Lots of flash crowds (content-driven behavior)– Lots of one-timers– Lots of diversity amongst clients, even for small
streams
![Page 35: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/35.jpg)
35
Nice pictures…see paper for details
0
20
40
60
80
100
rtp
http
mms
rtsp
Quicktime Real Windows media
Per
cent
age
of r
eque
sts
![Page 36: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/36.jpg)
36
Abandon all hope, ye who enter here.
![Page 37: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/37.jpg)
37
Talk outline
• Akamai live streaming workload• With an application end-point architecture
– Is the overlay stable enough despite dynamic participation?
– Is there enough upstream bandwidth?– Are overlay structures efficient?
• Summary
![Page 38: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/38.jpg)
38
When is a tree stable?
Not stable More stable
• Departing hosts have no descendants
• Stable nodes at the top of the tree
X
XX
Stable nodes
Less stablenodes
Interruptions
TimeAncestor leaves
![Page 39: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/39.jpg)
39
Extreme group dynamics
45% stay less than 2 minutes!
15% stay longer than 30 minutes(heavy-tailed)
![Page 40: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/40.jpg)
40
Stability evaluation: simulation
• Hosts construct an overlay amongst themselves using a single-tree protocol– Skeleton protocol of the one presented in the ESM
Usenix ’04 paper• Findings are applicable to many protocols
– Goal: construct a stable tree• Parent selection is key
• Group dynamics from Akamai traces (join/leave)
• Honor upstream bandwidth constraints– Assign degree based on bandwidth estimation
![Page 41: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/41.jpg)
41
Join
IP1IP2...
Join
![Page 42: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/42.jpg)
42
Probe and select parent
IP1
IP2
...
IP1
IP2
![Page 43: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/43.jpg)
43
Probe and select parent
• Oracle: pick a parent who will leave after me • Random • Minimum depth (select one out of 100 random)• Longest-first (select one out of 100 random)
Parent selection algorithms
![Page 44: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/44.jpg)
44
Parent leave
Host leaves
X
![Page 45: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/45.jpg)
45
Parent leave
Host leaves
All descendants are disconnected
?
![Page 46: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/46.jpg)
46
Find new parent
Host leaves
All descendants are disconnected
All descendants probe to find new parents
![Page 47: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/47.jpg)
47
Stability metrics
• Mean interval between ancestor change
• Number of descendants of a departing host
X
Interruptions
TimeAncestor leaves
![Page 48: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/48.jpg)
48
Stability of largest stream
Oracle: there is stability!
Min depth
Random
Longest-first
![Page 49: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/49.jpg)
49
Min depth, 82%
Random, 72%
Longest-first, 91%
Oracle, ~100% no descendants
Is longest-first giving poor predictions?
![Page 50: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/50.jpg)
50
Stability of 50 large-scale streams
Min depthRandom
Longest-first
OracleThere is stability! Of the practical algorithms, min depth performs the best.
Per
cent
age
of s
essi
ons
with
inte
rval
betw
een
ance
stor
cha
nge
< 5
min
utes
![Page 51: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/51.jpg)
51
There is inherent stability
• Given future knowledge, stable trees can be constructed
• In many scenarios, practical algorithms can construct stable trees – Minimum depth is robust– Predicting stability (longest-first) is not always
robust; when wrong, the penalty is severe
• Mechanisms to cope with interrupts are useful– Multiple trees (see paper for details)
![Page 52: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/52.jpg)
52
Talk outline
• Akamai live streaming workload• With an application end-point architecture
– Is the overlay stable enough despite dynamic participation?
– Is there enough upstream bandwidth?– Are overlay structures efficient?
• Summary
![Page 53: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/53.jpg)
53
Is there enough upstream bandwidth to support all hosts?
What if application end-points are all DSL?
Video 300 kbps
Upstream bandwidth only 128 kbps
DSLDSL
Saturated tree
DSL
![Page 54: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/54.jpg)
54
Metric: Resource index
• Ratio of the supply to the demand of upstream bandwidth Resource index == 1 means the system is saturated
• Resource index == 2 means the system can support two times the current members in the system
Resource Index:
(3+5)/3 = 2.7
![Page 55: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/55.jpg)
55
Large-scale video streams
Most streams have sufficient upstream bandwidth.
A few streams are in trouble or close.
1/3 of the streams are in trouble.
![Page 56: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/56.jpg)
56
Talk outline
• Akamai live streaming workload• With an application end-point architecture
– Is the overlay stable enough despite dynamic participation?
– Is there enough upstream bandwidth?– Are overlay structures efficient?
• Summary
![Page 57: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/57.jpg)
57
Relative Delay Penalty (RDP)• How well does the overlay structure match the
underlying network topology?
RDP = Overlay distance
Direct unicast distance
US
USEurope
US
US Europe
50ms
50ms 50ms
20ms
Results are more promising than previous
studies using synthetic workloads and topologies.
![Page 58: 1 CPS 214 Computer Networks and Distributed Systems “Live” Video and Audio Streaming End System Multicast Analysis of Akamai Workload](https://reader035.vdocument.in/reader035/viewer/2022062421/56649e555503460f94b4c435/html5/thumbnails/58.jpg)
58
Summary
• Indications of the feasibility of application end-point architectures– The overlay can be stable despite dynamic participation– There often is enough upstream bandwidth– Overlay structures can be efficient
• Our findings can be generalized to other protocols• Future work: Validate through real deployment
– On-demand use of waypoints in End System Multicast– Attract large groups
Thank you!