Title: Towards Efficient Simulation of Large Scale P2P Networks
1Towards Efficient Simulation of Large Scale P2P
Networks
ITG-Fachgruppe 5.2.1Cooperating and Scalable
Networks Aachen, Ericsson, J. Sachs, 04.05.2006
2Cartography of P2P Architectures
- Two control functions in P2P systems
- Resource mediationwhere are files located
- Resource access controlwho may download a file
and when - Mapping of P2P architectures into architectural
space - pure P2P
- hybrid P2P
- classic client/server
- Identification of control objectives
PureP2P
P2P Cartography
hybrid P2P
3Peer-to-Peer Architectures
Information Seeker
Information Provider
Information Mediator
Structured P2P-Network (Chord)
4Basic Functions of P2P Networks
new node
peer informs its neighbors and bootstrap server
node failures detected by periodical updates or
not answered requests resilience
5Basic Functions of P2P Networks (contd.)
peer x wants to insert (key,data) using DHTs
key hash(data)
peer x searches for key
x
and asks its neighbors
which redirect requests
y sends data to x
6Basic Function of P2P Content Distribution
- Main feature is multiple source download.
- Peers issue several download requests for the
same file to multiple providing peers in
parallel. - Providing peers serve the requesting peers
simultaneously.
providingpeer
After successfully downloading a whole chunk, it
is provided to other peers.
downloading peer 1
index server 2
index server 1
providingpeer
providingpeer
7Features of P2P Systems and Their Implications
- Usually a large number of participating peers
- Large-scale a lot of nodes and even higher
number of resources need to be simulated - Peers may arbitrary join or leave
- Highly dynamic a lot of user created event (due
to churn, i.e. peers joining and leaving
arbitrarily, as key feature of P2P systems) - Cooperative working of peers and robust systems
- Complexity
- one event can cause a large number of events at
other peers, i.e. system events, due to
cooperation among peers - additionally periodic or provisional systems
event to cope with the self-organization of p2p
systems
Target of Workshop Focus on large-scale P2P
networks in order to consider key characteristics
(e.g. regarding churn for 100 peers reasonable?)
8Approach For Simulating Large-Scale P2P Systems
- System state has to be stored at simulation
machine - requires efficient data structures (e.g. calendar
queue) - How to model in order to reduce the number of
events? - Resource mediation might not require to model
bandwidth, only signalling delay - Resource exchange might not require to model
delay if large contents are exchanged requires
modelling of bandwidth - other performance influence factors packet loss,
moving users, - appropriate abstractions models for
investigated application - Clustering of peers to user groups...
- might allow parallel simulation of clusters
9Workshop in Würzburg
- Efficient Data Structures
- Andreas Binzenhöfer, Calendar Queue and Event
Design Algorithms - Jens Oberender, Modelling Resource Fragmentation
- Abstractions and Models
- Kolja Eger, Packet-based Simulation
- Gerald Kunzmann, Signaling in Voice/Video over IP
Systems - Daniel Schlosser, Tobias Hoßfeld, Periodic and
Market-Based Bandwidth Allocation - Parallel Simulation
- Ivan Dedinski, Parallel Discrete Event Simulation
10Talks Today
Hier könnte Ihr Name stehen !
11Periodic and Market-Based Bandwidth Allocation
ineDonkey Networks
- Tobias Hoßfeld, Daniel Schlosser
12Measurements of eDonkey Traffic
- Case-by-case measurements of eDonkey file-sharing
application in public GPRS/UMTS network - Multiple source download via GPRS
13eDonkey Data Exchange via UMTS
- UMTS upload restricts throughput
- UMTS download restricts throughput
- Max-min-fair share of available bandwidth is
observed - How to model the bandwidth allocation of
fair-share P2P file-sharing applications?
14Simulation of Fair-Share Bandwidth Allocation
- Events which influence the bandwidth allocation
are that a peer - starts the download of a file
- finishs a download
- goes offline while downloading
- continues downloading a file after joining the
network again - We consider eDonkey-like file-sharing networks
- Aim Modeling of bandwidth allocation in
fair-share networks
Events
t
Dt
Dt
Dt
Dt
Dt
Dt
Dt
Dt
Dt
Dt
Dt
Computation of allocated bandwidth
t
15Stream-based or packet-based approach?
- TCP can be neglected if conditions are fulfilled
(540 kB blocks) - Signaling vs. data exchange RTT vs. bandwidth
16What means fair-share?
- All peers get the same bandwidth
- If a peer cannot consume completely the allocated
bandwidth, the surplus is distributed among the
remaining peers
3 kbps
11 kbps
13 kbps
13 kbps
17Periodic Bandwidth Allocation
- For each Dt, for each peer compute bandwidth
allocation
Allocated bandwidth can be overbooked or
underbooked
Peer 1
18Market-Based Bandwidth Allocation
- For each event, consider affected, i.e.
connected, peers - All affected peers make a bid
- Strategy
- If there are no other bids, propose x not
allocated bandwidth / peers - If minimal bid y of all affected peers is smaller
than x, then keep bid y and compute x - If all bids are larger than x, then bid x on
these connections - Finish If lower bid of a connection is the
minimal bid of a peer and is repeated
19Market-Based Bandwidth Allocation
Downloading network links
Uploading network links
20
NL3 40kbps
3.333
15
NL4 30kbps
NL0 10kbps
10
NL1 10kbps
5
NL5 10kbps
5
NL2 80kbps
NL6 10kbps
20
10
NL7 10kbps
Initial bid x BW / peers
40
NL8 40kbps
20MBBA Example
Downloading network links
Uploading network links
20
NL3 40kbps
3.333
15
NL4 30kbps
3.333
NL0 10kbps
26.667
10
NL1 10kbps
5
NL5 10kbps
5
3.333
25
NL2 80kbps
5
NL6 10kbps
25
6.667
25
20
10
NL7 10kbps
If minimal bid y of all connected peers holds
yltx, then set bid y and compute x x BW /
peers. If all yltx, keep x.
40
NL8 40kbps
21MBBA Example
Downloading network links
Uploading network links
20
NL3 40kbps
3.333
NL4 30kbps
3.333
NL0 10kbps
26.667
10
NL1 10kbps
5
NL5 10kbps
3.333
25
NL2 80kbps
5
NL6 10kbps
25
6.667
25
10
NL7 10kbps
Finish If lower bid of a connection is the
minimal of a peer and is repeated
40
NL8 40kbps
22MBBA Example
Downloading network links
Uploading network links
NL3 40kbps
3.333
NL4 30kbps
NL0 10kbps
5
26.667
NL1 10kbps
NL5 10kbps
24.444
NL2 80kbps
NL6 10kbps
6.667
6.667
24.444
24.444
10
NL7 10kbps
Finish If lower bid of a connection is the
minimal of a peer and is repeated Minimal bid
yltx, set y and compute x
40
NL8 40kbps
23MBBA Example
Downloading network links
Uploading network links
NL3 40kbps
3.333
NL4 30kbps
NL0 10kbps
5
26.667
NL1 10kbps
NL5 10kbps
24.444
31.667
6.667
NL2 80kbps
NL6 10kbps
10
24.444
24.444
31.667
10
NL7 10kbps
Finish If lower bid of a connection is the
minimal of a peer and is repeated If all yltx,
keep x.
40
NL8 40kbps
24MBBA Example
Downloading network links
Uploading network links
NL3 40kbps
3.333
NL4 30kbps
NL0 10kbps
5
26.667
NL1 10kbps
NL5 10kbps
6.667
NL2 80kbps
NL6 10kbps
31.667
36.666
10
NL7 10kbps
Finish If lower bid of a connection is the
minimal of a peer and is repeated
40
NL8 40kbps
25MBBA Example
Downloading network links
Uploading network links
NL3 40kbps
3.333
NL4 30kbps
NL0 10kbps
5
26.667
NL1 10kbps
NL5 10kbps
6.667
NL2 80kbps
NL6 10kbps
10
NL7 10kbps
36.666
NL8 40kbps
26Comparison PBA vs. MBBA
27Comparison PBA vs. MBBA
MBBA computes and allocates immediately
fair-share bandwidth -gt in real systems this
requires some time PBA bandwidth overbooked or
underbooked for ?t -gt in next step bandwidth is
adapted
28Conclusion
- For P2P content distribution networks, like
eDonkey, resource access control is crucial point - Fair-share bandwidth allocation has to be modeled
- We have proposed two stream-based approaches
which are valid in the considered scenarios - Periodic bandwidth allocation PBA
- Market-based bandwidth allocation MBBA
- Depending on the number of events influencing the
resource access control PBA or MBBA has to be
preferred