24
Multicast for Everyone Else: It’s finally here! Lenny Giuliano <[email protected]> Andrew Gallo <[email protected]>

Multicast for Everyone Else: It’s finally here! · 2019. 12. 9. · What’s old is new: Live streaming is Trending •Brand Transparency and Authenticity: due to spontaneous and

  • Upload
    others

  • View
    0

  • Download
    0

Embed Size (px)

Citation preview

  • Multicast for Everyone Else: It’s finally here!Lenny Giuliano

    Andrew Gallo

  • What is Multicast for Everyone Else (MFEE)?

    • Multicast ubiquitously available for EVERYONE (not just those in this room)!• Not just a hope for a distant future, but works today!

    2

  • Why Multicast?

    • First, some simple facts:• HD Bitrate: 4-8 Mbps • 4K Bitrate: 35-45 Mbps • 8K Bitrate: 80-100 Mbps• VR Bitrate: 500Mbps-5.2 Gbps!!!

    • Akamai Traffic Record (Dec 2018): 72 Tbps• Akamai estimated to deliver 15-30% of total web traffic• 72Tbps / 40Mbps = 1.8m simultaneous 4k viewers to set new traffic record

    3

    https://datatracker.ietf.org/meeting/105/materials/slides-105-mops-mops-taxonomy-of-issues

  • #179/207 “The Resident”: 1.848m

    * Source: Jake Holland, Akamai

    https://datatracker.ietf.org/meeting/105/materials/slides-105-mops-mops-taxonomy-of-issues

  • Can Brute Force Unicast (BFU) Keep Up??

    • TV Audience of average NFL game: ~15M• 15M viewers @ 40Mbps = 600Tbps!!

    • TV Audience of other major events:• Super Bowl 100m• 2015 Cricket World Cup (India played): 200-300m• 2018 FIFA World Cup Finals: 500m

    • “Cord Cutting” Evolution- live, linear TV is the last frontier• YouTubeTV, SlingTV, HuluTV, etc

    • Live, linear TV is not dead yet!

    5

  • What’s old is new: Live streaming is Trending

    • Brand Transparency and Authenticity: due to spontaneous and un-editable nature, live streams are perceived as more authentic and drive greater emotional engagement than on-demand

    • Viewers spend 8X longer with live video than on-demand (5.1 minutes for on-demand vs. 42.8 minutes for live video content)

    • 67% of live video viewers are more likely to make a purchase• >50% of marketing professionals said they have seen the best ROI from live video than

    any other social media platform• Live video is outpacing the growth of other types of online video, with a 113% increase

    in ad growth yearly

    6

    https://50wheel.com/the-top-video-live-streaming-statistics-to-know-in-2018/https://50wheel.com/the-top-video-live-streaming-statistics-to-know-in-2018/http://www.mediative.com/2018-digital-trends-10-live-streaming-video/https://admarketing.yahoo.net/rs/118-OEW-181/images/Yahoo_The%2520Live%2520Video%2520Opportunity_2016.pdf

  • • Any argument for/against IPv6 applies just as well to Internet Mcast• It’s not needed: NAT, CDN• It is needed: v4 address exhaustion, inefficiency of duplicated video streams

    • Both suffer the “all or nothing” problem• Both require new protocols/extensions of old ones• Both born in the 90s, struggled through adolescence, un/underemployed as

    young adults• Both require faith in the unseen• Both do accomplish little on their own, but enable big things

    Internet Mcast and IPv6: Technological Cousins

  • Conclusion: Live linear video is here to stay

    • … but, IP (unicast) is not well suited to multi-destination traffic• IP Multicast was created 2+ decades ago to solve this problem• So, what went wrong?

    8

  • In the beginning, there was unicast…$$$$!!!

  • …then came Multicast

  • But, there was a problem with multicast…Internet

    Mcast-enabledMcast-enabled

    Mcast-enabled Unicast-only

    Unicast-onlyUnicast-only

    Unicast-only

    Unicast-only

    • The “All or Nothing” problem

  • What to do? CDNs!• Content Delivery Networks became popular• Almost as good as multicast, and it just works!

  • Comparison: Unicast, CDN and Multicast

    Audience

    Cost

    Unicast

    Multicast

    CDN

    • Cost (audience) for the Content Provider• CDNs don’t eliminate the BFU problem, they just distribute it

  • So What’s New?• Internet Multicast has been around since the 90’s and hasn’t gained much traction, so

    why will this time be different?

    • ASM Deprecation- SSM-only, eliminates most of the complexity of multicast• draft-ietf-mboned-deprecate-interdomain at IESG now, soon to be RFC• BCP already applied in I2

    • Automatic Multicast Tunneling (RFC7450)• Addresses fundamental “all or nothing” problem of multicast• Tunneling is fashionable- “Overlay Networking” is cool now!

  • AMT- How it works

    Mcast Enabled ISP

    Unicast-Only Network

    Content Owner

    Mcast Enabled Local Provider

    Mcast Traffic

    Mcast Join

    UDP-IGMP Request

    •Gateway (client) uses Anycastto discover nearest relay.•UDP-encapped IGMP report is sent to relay at the edge of the mcast network.

    Relay

    GW

  • AMT – Unicast Edge NetworkMcast Enabled ISP

    Unicast-Only Network

    Content Owner

    Mcast Enabled Local Provider

    Mcast Traffic

    Mcast Join

    UDP-IGMP Request

    Mcast join is sent by relay on behalf of the unicast receiver.

    Relay

    GW

  • Mcast Enabled ISP

    Unicast-Only Network

    Content Owner

    Mcast Enabled Local Provider

    AMT – Unicast Edge NetworkMcast Enabled ISP

    Unicast-Only Network

    Content Owner

    Mcast Enabled Local Provider

    Mcast Traffic

    Mcast Join

    UDP-IGMP Request

    Mcast stream is replicated and encapped in UDP to the unicast receiver.

    UDP Ucast Stream

    Relay

    GW

  • Mcast Enabled ISP

    Unicast-Only Network

    Content Owner

    Mcast Enabled Local Provider

    AMT – Unicast Edge NetworkMcast Enabled ISP

    Unicast-Only Network

    Content Owner

    Mcast Enabled Local Provider

    Mcast Traffic

    Mcast Join

    UDP-IGMP Request

    UDP Ucast Stream

    Buy a bigger pipe, or enable multicast, or both?(AMT is a stopover, not a destination)

    RelayGW

    GWGW

  • Steps for World DominationMFEE in 5 steps:• Step 1: AMT Relays

    • 1st Relay deployed at TJHS by William Zhang for senior project• Two more relays deployed at GWU

    • Step 2: AMT GW implementations• Wanted in-browser implementation, but not possible today (*)• GW implemented in VLC by Natalie Landsberg/Wayne Brassem, available in 4.0.0

    • Step 3: Portal for mContent• William wrote script to find all active sources (non-beacons) on I2 Looking Glass (found 119 sources > 5pps) • Andrew created VLC Playlist from script

    • Step 4: Off-Net Sourcing• Simple way for end users to add live multicast content to the MBone• Successfully demo’d with iPhone (ping) at IETF Singapore using LISP; working on camera video…

    • Step 5: ???19

    Profit!!

  • MFEE Architecture: Putting it all togetherInternet

    MBONE (I2)

    Mcast Source

    Relay

    Off-net Receiver

    Off-net Source LISP ETR

    AMT

    LISP

    RelayRelay

    Native Receiver

  • How to get involved

    Lots more to be done:• In-browser implementation (kicked off at IETF)• Portal- curation script to populate “TV Guide” of

    active streams• Add more relays• Off-Net Sourcing- need help getting phone camera

    to transmit multicast packets• Grad students wanted!• Content, content, content!• Tshirts available…• See me at the Juniper booth (or bar) if interested

    21

    Multicast

  • Why Internet2?

    • The MBONE is synonymous with Internet2• Is there are larger multicast network around?• One that has lots of inter-domain traffic?

    • Internet2 is both a testbed and production network.• AMT is a production-ready technology, but needs exposure

    • We’re a community that likes to try new things the commercial guys wont play with

    • AMT democratizes and decentralizes content sourcing, something the internet sorely needs in 2020

    22

  • Wanna play?• STEP 1: Download version of VLC with AMT Integration

    (any nightly build after 11/11/19)• https://nightlies.videolan.org/

    • STEP 2: See details for usage• See README at https://github.com/Juniper/amt-vlc

    • STEP 3: Open some streams:

    23

    Content URL

    Big Buck Bunny amt://[email protected]

    Elephant's Dream amt://[email protected]

    Sintel amt://[email protected]

    NASA amt://[email protected]

    Science content from GMU amt://[email protected]

    (your content here) Please share interesting multicast sources

    https://nightlies.videolan.org/https://github.com/Juniper/amt-vlc

  • Acknowledgements and Further Reading• Pete Morasca (TJHS)• William Zhang (TJHS, UVa)• Natalie Landsberg/Wayne Brassem (Juniper)• Dino Farinacci (lispers.net)• References

    Natalie/Wayne’s VLC with AMT GW buildhttps://github.com/Juniper/amt-vlc

    William’s Senior Project Docshttps://github.com/willzhang05/senior-research

    Curation Resultshttps://github.com/willzhang05/senior-research/blob/master/logs/results.txt

    • Lots more technical details to cover• Join the team at mttg.slack.com, send us email to join ([email protected], [email protected])

    https://github.com/Juniper/amt-vlchttps://github.com/willzhang05/senior-researchhttps://github.com/willzhang05/senior-research/blob/master/logs/results.txt