mario gerla current network research projects ad hoc, wireless networks (darpa, nsf, onr) wireless,...
TRANSCRIPT
Mario GerlaCurrent Network Research Projects
• Ad hoc, wireless networks (DARPA, NSF, ONR)• Wireless, mobile access to Internet (NSF, Intel)
• Internet : QoS Routing and multicasting (CISCO, NASA, NSF)
• Internet control models: TCP (EPRI,NASA)
• Internet II: high speed traffic models and measurements (NSF, EPRI)
www.cs.ucla.edu/NRL
Cellular Vs Multihop
Ad Hoc, Multihop wireless Networks
Base BaseBase
Standard Base-Station Cellular Networks
Challenging problem: multihop routing
• mobility• need to scale to large numbers (100’s to 1000's)• unreliable radio channel (fading etc)• limited bandwidth• limited power• need to support multimedia (QoS)
Conventional routing: Distance Vector
0
5
1
2
4
3
Destination Next Hop Distance
0 2 31 2 2… … …
Routing table at node 5 :
Conventional wired routing limitations
• Distance Vector (eg, Bellman-Ford, DSDV):– routing control O/H linearly increasing with net size
– convergence problems (count to infinity); potential loops
CONVENTIONAL ROUTING DOES NOT SCALE TO SIZE AND MOBILITY
Fisheye State Routing
• Routing information is periodically exchanged with neighbors (as in Distance Vector)
• BUT: Routing update frequency decreases with distance to destination – Higher frequency updates within a small radius and lower frequency
updates to remote destinations
– Result: Highly accurate routing information about immediate neighborhood; progressively less detail for areas further away
Scope of Fisheye
1
2
3
4
5
67
8
9
9
10
11
12
14 1516 17
18 19
20
21
2223
2425
26
27
28
29
30
31
3234
35
36
Hop=1
Hop=2
Hop>2
13
How to deal with remote destination inaccuracy? Landmark Routing
LandmarkLandmark
Logical SubnetLogical Subnet
Snapshot
A
B
C D
HI
JK L
O
P
LM1
LM2
LM3
LM4
GloMoSim Simulation Layers
Application Processing
Propagation Model Mobility
Frame Processing Radio Status/Setup
CS/Radio SetupRTS/CTSFrame Wrapper
Ack/Flow Control
Clustering
Packet Store/Forward VC Handle
FlowControl Routing
IP Wrapper IP/Mobile IP
RSVPTransport Wrapper TCP/UDP Control
Channel
Radio
MAC Layer
Network
IP
Transport
Application
RTP Wrapper RCTP
Packet Store/Forward
Clustering
Routing
Link Layer
Application Setup
Data PlaneData Plane Control PlaneControl Plane
Ad Hoc, Personal Networking with Bluetooth
headset
cell phone
storage
palmtop
PDA
What Is Bluetooth?
Personal Ad-hoc Personal Ad-hoc NetworksNetworks
Cable Cable ReplacementReplacement
Landline
Data/Voice Data/Voice Access PointsAccess Points
Wireless Network
UCLA Adaptive Speech Experiment
Multihop Testbed
client
• Adjustable Parameters - sampling rate - packet size
• QoS Monitoring: - packet loss - jitter
Audio(UDP)
Control(TCP)
A d a p t a t I o n S t r a t e g y :
Audio source adapts to QoS feedback
Increase in Packet loss packet size is reduced
sampling rate is reducedIncrease in jitter network congested
channel noise/interference
Piggybacked Text Stream(UDP)
server
TTSSync
SpeechRecognition
iMASH: Interactive Mobile Application Support for Heterogeneous clients
CS: R. Bagrodia, M. Gerla, S. Lu, L. Zhang
Medical School: D. Valentino, M. McCoy
Campus Admin: A. Solomon
UCLA
Supported by NSF
Diverse Display Devices
Use of different devices for different components of medical care
Imaging Workstation: high-quality medical imagery and multimedia patient records
Medical Workstation: multimedia patient records, including moderate-resolution images
Mobile Medical Notes: for reviewing and taking medical notes
Physician’s PDA: for messaging and scheduling
Hardware & Connectivity
ApplicationServer
High bandwidthIntranet
MiddlewareServers
MiddlewareServers
MiddlewareServers
iMASH: Components
• Target application: Mobile physicians
• Middleware infrastructure to support anytime, anywhere, any-device access to electronic multimedia data
• Protocols to provide reliable QoS in a mobile, heterogeneous network
• Simulation/emulation capability to evaluate scalability of system to many users over large geographic areas
• Limited evaluation via deployment within UCLA medical school
QoS Routing and Multicast in wired nets
• Supported by CISCO and by NASA AMES• Intradomain environment• Quality of Service Routing/Multicast for
Real Time traffic (IP telephony,video etc)• Call Admission Control• Traffic load balancing
Example of QoS Routing
A
B
D = 30, BW = 20D = 25, BW = 55
D = 5, BW = 90
D = 3, BW = 105
D =
5, B
W =
90
D = 1, BW = 90
D = 5, B
W = 90
D =
2, B
W =
90
D = 5, BW = 90D = 14, BW = 90
Constraints: Delay (D) <= 25, Available Bandwidth (BW) >= 30
Multiple constraints QoS Routing
Given:
- a (real time) connection request with specified QoS requirements (e.g., Bdw, Delay, Jitter, packet loss, path reliability etc); examples: IP telephony, video streaming
Find:
- a min cost (typically min hop) path which satisfies such constraints
- if no feasible path found, reject the connection
2 Hop Path --------------> Fails (Total delay = 55 > 25 and Min. BW = 20 < 30)3 Hop Path ----------> Succeeds!! (Total delay = 24 < 25, and Min. BW = 90 > 30)5 Hop Path ----------> Do not consider, although (Total Delay = 16 < 25, Min. BW = 90 > 30)
A
B
D = 30, BW = 20D = 25, BW = 55
D = 5, BW = 90
D = 3, BW = 105
D =
5, B
W =
90
D = 1, BW = 90
D = 5, B
W = 90
D =
2, B
W =
90
D = 5, BW = 90D = 14, BW = 90
Constraints: Delay (D) <= 25, Available Bandwidth (BW) >= 30
We look for feasible path with least number of hops
Benefits of QoS Routing
Without QoS routing: • must probe path & backtrack; non optimal path, control traffic
and processing OH, latency
With QoS routing:• optimal route; “focused congestion” avoidance• more efficient Call Admission Control (at the source)• more efficient bandwidth allocation (per traffic class)• resource renegotiation possible
High Speed Networks Performance Measurement and Analysis
Mario Gerla and Medy Sanadidi
Project Focus
• High speed : backbone links at 2.4 Gbps and above, as in Abilene and vBNS
• Heterogeneous networks: wired and wireless• High performance distributed applications:
processor intensive, large data bases, high traffic volume, low latency
• Application performance : measure the network performance as perceived by network applications/users; tune protocols to improve performance
Example: Urban Simulation(R. Muntz and B. Jepson)
• Real-time visual simulation for design, urban planning, emergency response, and education
• Built Virtual Los Angeles model
• Challenge: remote/distributed access through high speed net
Current Measurement Activities
• TCP performance over wireless Internet access links (wireless LAN, satellite); wireless, lossy channel emulator; TCP Westwood
• Characterization of long range dependent traffic in the Internet; traffic generators
• Measure performance of dataView (3 D rendering of scientific data): impact of propagation time and network bottlenecks