Connectivity for Advanced VoIP and Integrated Communications

1
Connectivity for Advanced VoIP and Integrated
Communications

• National Internet2 DayMarch 18th, 2004
• Ben Teitelbaum, Internet2
• For this and other talks...
• http//www.internet2.edu/ben/talks.html

2
Personal Communications

• Highest value applications are generic
• Email, telephone, postal mail
• Throughout the history of communications, generic
  person-to-person communications have dominated
  content distribution and domain-specific apps
• Voice
• The dominant real-time communications medium
• A "killer" network application for 100 years
• VoIP opens doors to make voice communications
  even more valuable

3
Voice / Data Convergence

• Long promised, but still inevitable
• Has happened in the core
• Happening now for residential / enterprise
• Drivers
• Cost, cost, cost, cost, cost
• Why have a giant switch, two networks, two
  staffs, etc.?
• Voice is just another network application
• Dampers
• QoS, 911, CALEA, USF, FUD
• Little user "pull"

4
Advanced Voice

• Replace or Displace?
• If voice is so important, why be satisfied with
  replacing plain-old telephone service (POTS)?!
• Users are already defecting to cellular, SMS,
  AIM, MSN Messenger, Yahoo!, etc.
• VoIP can be much better than POTS
• Potential dimensions of improvement
• Fidelity
• Privacy
• Addressing
• Mobility
• Media translation

• Survivability
• Integration with IM, video, etc.
• Presence
• "911"

5
Advanced Voice

• Replace or Displace?
• If voice is so important, why be satisfied with
  replacing plain-old telephone service (POTS)?
• Users are already defecting to cellular, SMS,
  AIM, MSN Messenger, Yahoo!, etc.
• VoIP can be much better than POTS
• Potential dimensions of improvement
• Fidelity
• Privacy
• Addressing
• Mobility
• Media translation

• Survivability
• Integration with IM, video, etc.
• Presence
• "911"

Not covered here, but Internet2 projects in these
areas exist
Covered in this talk
6
Connectivity

• "Paths in the snow"
• Don't predict how users will want to communicate
• Users are highly-motivated to communicate with
  each other and to innovate
• Connect them and watch what happens
• Witness P2P file-sharing
• Network connectivity middleware P2P FS
• Similar potential for real-time apps
• We should get ahead of the curve this time and
  provide the enabling middleware ourselves

7
Connective Middleware
User
User
8
Internet2's Secret Sauce

• Demographics
• 3.8 million students (tech-savvy, talk a lot,
  adapt easily)
• And, by the way, they graduate (tech-transfer à
  la email)
• Institutional Commitments
• Internet2 members have committed to advance IP
  communications and promote collaborative apps
• Many are looking for ways to reverse eroding
  voice revenues
• Connectivity
• Great networking connectivity
• High-bandwidth, low-loss, low-jitter
• End-to-end transparency (few NATs)
• IPv6 and multicast too!
• Emerging middleware infrastructure for AuthN/Z
• Need to build on this to connect users with each
  other!

9
SIP.edu
Whats SIP?
10
SIP Session Initiation Protocol (RFC 3261)

• SIP
• Signalling protocol for creating, modifying, and
  terminating real-time internet media sessions
• SIP (and its extensions) support traditional
  telephony features as well as instant messaging
  and presence
• Attributes
• Provides user registration, call routing, setup,
  tear down, and redirection
• Makes heavy use of existing standards
• Easy and familiar feel (email-style headers,
  HTTP-style error codes, URL addresses)
• Signaling and media paths separate

11
SIP Trapezoid
Proxy
Proxy
Registrar
Registrar
Alice's hardphone
Bob's softphone
Media (and subsequent signaling)
12
SIP Happens

• Vigorous standards activity
• SIP (core protocol and extensions) SIPPING
  (applications) SIMPLE (instant messaging and
  presence) not just IETF, 3GPP too!
• Enthusiastic industry adoption
• Phones (Microsoft, Cisco, Pingtel, Snom, ...)
• Servers (Cisco, Microsoft, Broadsoft, ...)
• Conferencing (eDial, RADVision, ...)
• Services (Level3, WorldCom, Vonage, ...)
• Open source software
• Servers SER (iptel.org) VOCAL (vovida.org)
• Soft Phones Linphone KPhone

13
SIP.edu

• Fearless Leader
• Dennis Baron
• Goals
• Grow number of SIP connected users
• Increase value proposition for end-user SIP
  adoption
• Promote convergence of voice and email identity
• Low entry-cost means for campuses to...
• Provide a useful service
• Start getting their feet wet with SIP
• Means
• SIP.edu cookbook available on web site
• Partnering with vendors (Cisco working with 6
  schools)

14
Remember It's People We Are Connecting

• Addressing
• Users should not be burdened with device
  addresses, when its people they really care
  about
• Addresses should be mnemonic and empower
  enterprises to manage the identities of their
  users
• sipben_at_internet2.edu
• Its time to put E.164 phonenumbers behind us!
• A.G. Bell did not say...

15
SIP.edu Architecture (today)
SIP User Agent
INVITE(sipbob_at_bigu.edu)
DNS SRV query sip.udp.bigu.edu
bigu.edu
PRI / CAS
INVITE(sip12345_at_gw.bigu.edu)
sip. udp.bigu.edu IN SRV ...
telephoneNumberwhere mailbob
Bob's Phone
16
SIP.edu Architecture (real soon)
SIP User Agent
INVITE (sipbob_at_bigu.edu)
DNS SRV query sip.udp.bigu.edu
bigu.edu
INVITE (sipbob_at_207.75.164.131)
REGISTER(Contact 207.75.164.131)
Bob's SIP Phones
17
SIP.edu Growth
8 other schools working on it
18
Presence and Integrated Communications

• Presence
• Notification of events that facilitate
  communication (Henning Schulzrinne)
• On-line, Away, Idle, On phone, Out to
  lunch, ...
• Back to the future?
• Remember finger, write, who?
• Presence restores the sense of community that
  existed on timesharing systems
• Forward to the future!
• New standards for interoperability and
  scalability
• User-centric control of presence publication
• Richer state semantics and automatic triggers

19
Rich Presence Trials
Indianapolis October, 2003 Honolulu January,
2004 Arlington April, 2004
20
Client Mockup

• WaveThree Software and Columbia U. have provided
  clients (Session, SIPC)
• Others welcome! (client requirements doc on web)

21
VoIP Working Group

• Chairs
• Walt Magnussen
• Mike Enyeart
• Web site
• http//voip.internet2.edu/
• Projects
• SIP.edu
• Voice Disaster Recovery
• H.323 VoIP Testbed

22
Presence and Integrated Communications Working
Group

• Chair
• Jermey George
• Web site
• http//pic.internet2.edu/
• Projects
• Rich Presence Trials
• Social Context Study Group

23
(No Transcript)