Transcript kickoff0598 - BNRG - University of California, Berkeley

ICEBERG:
Internet core for CEllular
networks BEyond the thiRd
Generation
Bridge to the
Future
A research project
at the University of
California, Berkeley
Randy H. Katz
Anthony D. Joseph
Cellular “Core” Network
Emerging Distributed System
Architecture Spanning
Processing and Access
Personal Information Management and “Smart Spaces”
Distributed Videoconferencing
Room-scale Collaboration
Speech and Location
Aware Applications
ICEBERG
Computer-Telephony Services
MASH Media Processing Services
Active Services Architecture
TranSend Extensible
Proxy Services
Distributed Computing Services: Ninja
Computing and Communications Platform: Millennium/NOW
MASH Active Services
• Services supported within the network
– Service deployment/instantiation
» Service discovery service
– Service extension/customization
» JAVA or C++ programming model vs. TACC
– Service execution “platform” based on clusters
» Service Agent (servent): instance in execution
» Scalability, robustness based on soft-state, announce-listen
• 1st instance: Elan Amir’s video transcoding proxies
– Manage video streams across bottleneck links
• 2nd instance: Angie Schuett’s video archive server
– Introduce stateful services, more complex service deployment
MASH Active Services
• Platform Architecture
– Cluster computing
– Multicast-based announcelisten protocols
• Platform Management
– Resource management
– Load balancing, servent
scaling to offered load,
robust keep-alive
mechanisms
• Service Environment
– Defining services via
(MASH shell) scripts
• Service Management
– Launching and halting
servents
– Decentralized via
announce-listen + MC
damping
• Platform Location
– Locating a service
• Service Composition
– Servent interaction to
enable processing
pipelines
– E.g., servent client of
another servent
• Service Control
– Control protocols running
between clients and
servents
NINJA Infrastructure
• Focus on component services
Vertically “integrated” services
Component Services
E.g., dynamic composition, rapid deployment, reuse, data only, UI defined dynamically based on
device/connection, competition at every level …
Units (end devices), Active Routers (soft-state), Bases (persistent state)
Operators, typed connectors, and paths
Experimental Testbed
Fax
IBM
WorkPad
Image/OCR
Text
Speech
MC-16
Ericsson
CF788
306 Soda
Motorola
Pagewriter 2000
WLAN
405 Soda
326 Soda “Colab”
Pager
GSM BTS
Network
Infrastructure
Millennium Cluster
Smart Spaces
Personal Information Management
Millennium Cluster
Personal Information
Management
Speech-to-Voice Mail
Speech-to-Voice Attached-Email
Call-to-Pager/Email Notification
Email-to-Speech
All compositions
of the above!
Universal In-box
Policy-based
Location-based
Activity-based
Industrial Sponsors:
Committed and Potential
• Ericsson GSM
basestation and
telephony handsets
• IBM Workpad thin client
access devices
• Lucent (GSM group and
Inferno OS groups)
• Motorola two-way
pagers
• Sun Microsystems
Network Appliances
Group
• ATT Internet Lab (Geoplex
Menlo Park)
• Microsoft
• Intel Home Networking
Group
• Xerox
• Sprint
• 3COM
ICEBERG Project Vision
• Third Generation Cellular Architectures:
– Will support diverse air interfaces with different coverage,
bandwidth, latency characteristics
» TDMA, CDMA, wide-area, local-area, satellite, etc.
– Segregated circuit-switching for voice and packetswitching for data (e.g., GPRS)
• We Will Go Beyond the Third Generation:
– A lower cost, more flexible core network can be built using
full packet-switching techniques
– Delay sensitive and delay insensitive flows are easier to
support at the same time in a full packet-switching
architecture
– Processing embedded in the network enables more rapid
deployment of new kinds of applications and services
ICEBERG Project Goals
• Exploit Expertise in IP Protocol Suite and Proxy
Architectures to
– Demonstrate ease of new service deployment
» Packet voice for computer-telephony integration
» Speech- and location-enabled applications
» Complete interoperation of speech, text, fax/image across the
four P’s: PDAs, pads, pagers, phones)
» Mobility and generalized routing redirection
– Demonstrate new system architecture to support innovative
applications
» Personal Information Management
• Universal In-box: e-mail, news, fax, voice mail
• Notification redirection: e.g., e-mail, pager
» Home networking and control of “smart” spaces,
sensor/actuator integration
• Build on experience with Colab, 306/405 Soda
ICEBERG Project Goals
• Understand
– Implications for cellular network design based on IP technology
» IP network provisioning for scalability
» Pragmatic QoS for delay-sensitive flows
» Multinetwork mobility and security support
– How to use the emerging Ninja/Active Services infrastructure to
» Encapsulate existing applications services like speech-to-text
» Deploy and manage such computationally intensive services
in the network
» Integrate other kinds of services, like mobility and
redirection, inside the network
Project Strategy
GSM Infrastructure Elements
-- Data over PBMS GSM Network
-- GSM Base Station
-- Integration with IP-infrastructure
Analyze
Existing Systems
Prototype Elements
Design
-- Handset/computer integration
-- Java-enabled components
-- ProActive infrastructure
Next
Generation
ns Simulations
-- Ericsson channel error models
-- GSM-based infrastructure
-- GSM media access & link layer
Implement
New System
Specific ICEBERG Project
Areas
• Mobility Management
• Packet Scheduling in GPRS and W-CDMA
• Proxy- and Multicast-Enabled Services
Mobility Management
• Mobile IP-GSM Mobility Interworking
– Mobile IP-GSM authentication interworking
– Scalability of Mobile IP/hierarchical agents
• Multicast support for mobility
– Alternative approach for mobility based on M/C addresses
– Exploit multicast routing to reach mobile nodes without explicit
handoff
– Combine with real-time delivery of voice and video
• Generalized redirection agents
– Policy-based redirection: e.g., 1-800 service, email to pagers, etc.
– Redirection agents collocated with multicast tree branching
points
Packet Scheduling
• Validated ns modeling suite for GSM media
access, link layer, routing, and transport layers
– GSM channel error models
• QoS-aware High Speed Circuit Switched Data
(HSCSD), General Packet Radio System
(GPRS), and Wideband CDMA (W-CDMA) link
scheduling
–
–
–
–
RSVP signaling integration with bottleneck link scheduling
Fairness and utilization for TCP and RTP flows
Delay bound scheduling for R/T streams
Exploiting asymmetries in downstream/upstream slot
assignment, CDMA self-interference
New Services
• Proxies for Telephony-Computing Integration
– GSM-vat-RTP interworking: handset-computer integration
– Encapsulating complex data transformations
» Speech-to-text, text-to-speech
– Composition of services
» Voice mail-to-email, email-to-voice mail
– Location-aware information services
» E.g., traffic reports
– Multicast-enabled information services
» Multilayered multicast: increasing level of detail as
number of subscribed layers increase
Project Schedule
• Year 1: 1998
– ns modeling, validation
– GSM BTS-IP integration
– Initial design of mobility interworking and intelligent
networking services
• Year 2: 1999
– GSM-Wireless LAN integration
– Design of information-push applications
– Implement mobility interworking
• Year 3: 2000
– Extend testbed with W-CDMA and GPRS
– Roaming, scheduling, new applications demonstrations
– Fine-tuning and documentation
Goals for Today
• Review where we are
and determine who is
working on what:
– Infrastructure Building:
simulation models, testbed
elements
– Performance and
Modeling: data over
cellular, GPRS scheduling
– Network Design Issues
» Mobility
» Scalability
– Service Architecture and
Applications
• Realizable milestones
for the summer:
– 10 weeks between now
and August!
– Posters for June 10-12
BARWAN Retreat
» Excellent opportunity
for industry feedback
– Ericsson 2 Day Review in
Early August
» Initial results
» Demonstrations!