How you shouldn't just look at IP technologies in broadcast but also look at how off-the-shelf IT equipment can be used. Presented at NAB BEITC Engage! 2017
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...
Don't just go IP - Go IT
1. Don’t Just Go IP, Go IT!
Kieran Kunhya – kierank@obe.tv
@openbroadcastsy - #NABShow
2. What I’ll talk about
today• IT in Live Broadcast Television (orders of
magnitude differences for latency/bitrate
compared to OTT).
• History of IT in Live Broadcast
• IP/IT Technical challenges
• Using IT equipment in live broadcast
3. Who are we?
• Central London based company
• Build broadcast hardware and
software using IT equipment
• Written entirely in-house
• Video contribution and general
processing.
• Involved in early multivendor IP
deployments.
4. Traditional hardware architecture
• Single or limited function
cards with onboard
FPGAs or DSPs doing
majority of work.
• Smaller CPU to perform
control aspects.
5. A true software architecture
Standard CPU performs all non-
physical data processing
ASI
SDI
MADI
AES/EB
U
IP
etc.
ASI
SDI
MADI
AES/EB
U
IP
etc.
6. A true software architecture
Standard CPU performs all non-
physical data processing
ASI
SDI
MADI
AES/EB
U
IP
etc.
ASI
SDI
MADI
AES/EB
U
IP
etc.
7. Technological
changes• Intel Nehalem (2008/09)
• Realtime 1080i encoding
• Only possible with overclocking
before
• Low cost SDI/ASI boards
• Full 10-bit, VANC
• Commodity hardware
Everything available next-day!
8. Myth 1: IT hardware is too large and too
power hungry
Units reside in street
furniture outside Downing
Street, Buckingham Palace
etc. Low depth (200mm)
chassis, low power CPU.
Encode/decode, ASI
processing, Intercom.
9. Myth 2: IT hardware isn’t powerful enough
And another 17 more
cropped out!
• Huge processing power available
• 432 CPUs (small supercomputer)
• Costs falling dramatically
• Powerful enough for example to do high-
bitrate VC-2 encoding/decoding, very
high density H.264 decoding, ASI
processing.
• ~2-3x less rackspace for equivalent
functionality compared to hardware
• Processing power for new functions
10. Standards in software
contextSMPTE 2022-6 – SDI over IP
• Samples spanning packets (awkward)
• Costly CRC
• SDI frame tedious to construct
TR-03 – separate RTP streams for video/audio
• Samples no longer span packets (RFC4175)
• Soon SMPTE 2110
• Will this actually be possible in software?
11. Pixel formats
Only YUV 4:2:2 domain (as example)!
• Planar 10b – main working format
• Planar 8b - preview quality
• UYVY 10b (16-bit aligned) – SDI
datastream
• Apple v210 – hardware
• Contiguous 10-bit – 2022-6/TR-03 packing
Tricky to work
with in software.
12. Handwritten (no intrinsics!) SIMD for every mapping (and
others).
• 5-15x speed improvements compared to C
• Do it once, make it fast once and for all (until new
CPU…)
• Generic conversion library a difficult problem
• Intermediate pixel format(s) always a compromise
• Add special cases until you’ve done them all!
Pixel formats
13.
14. One does not
simply…Bypass the operating system
DPDK, Netmap, Registered I/O
A presentation in itself. See
BBC R&D at UKNOF:
https://www.youtube.com/watch?v=yL
L8wl8YUwA
15. Kernel Bypass
No network stack – You are the network stack
• Craft Ethernet, IP, UDP headers yourself
• No ARP, hardcoded MAC
• Handle most of this in userspace
• Lets you do pixel processing directly to/from
Network Card memory.
16. Packet Spacing
(1)
THE million dollar
question
• Left unspecified in SMPTE 2022-6
• Hardware can do this with ~ns precision (measured
with HFT equipment), but what about software?
17. Packet Spacing
(2)• Varies by NIC vendor, measured two, both bimodal
• Some implementations require “fake-packets”
• More research needed!
• Fractional Framerates 2.0 if
constraints too tight
18. IT-based live production
today
• Dual-capable, SDI/ASI/AES and IP
• Make it look the same to end-users, buttons,
SNMP etc.
• Incremental improvements like a web
browser
• Allow users to understand multifunctionality
• Integrate seamlessly (e.g MPEG-TS)
19. IT-based live production tomorrow
• A BIG COMPUTE RESOURCE
• Live and offline on the same CPU
• Remote kit can be processing low-priority batch
jobs
• Flash override on live broadcasts
• Possible today but no orchestrator
(microservices?)
• FIMS too old (SOAP/XML…)
• Control network and CPU resources
20. Decode
MPEG-TS
multicast
Orchestrator goes and finds
CPU resources somewhere,
onsite or otherwise.
Spin me up
a
multiviewer
Transcode
this file
Not possible
with SDI
because SDI
associated with
a connector on a
machine.
21. The V-word,
Virtualisation• OS provides priorities to run multiple applications
already
• Many vendors on Windows (often without good
reason)
• Getting > 1Gbps of data out of VM nontrivial.
• Live migration of complex, stateful broadcast
applications not easy. (Not a web server…)
• Complexity may outweigh benefits.
22. Using IT networks for contribution
• Trend towards using generic, unmanaged, IP networks for broadcast
contribution.
• Regular in News (including cellular)
• Growing use for Sport and Channel contribution
• Important for any Cloud workflows (e.g Amazon Web Services)
• Packet loss and Jitter (relative to traditional metrics)
• Do it properly, using UDP. TCP not suited.
• Variations on three main techniques.
23. Using IT networks for contribution
(1)• SMPTE 2022-1/2 FEC
• XOR based matrix (adds 2 * matrix latency)
• Basic but wide support (albeit many broken
implementations)
Row FEC Column FEC
24. Using IT networks for contribution
(2)• Retransmits (aka ARQ)
• Receiver requests
sender to transmit a
copy of lost packet.
• Affected by round-trip
latency
• Negative
acknowledgment
Sender Receive
r
25. Using IT networks for contribution
(3)• Dual-pathing (SMPTE 2022-7)
• Hitless Switching
Path1
Path2
Out
26. Example of IT and Generic IP at
Scale• Deluxe Broadcast Delivery
Network
• Using software-based
infrastructure for encoding,
delivery and decoding.
• Transported using multiple IP
connections
• Delivered NFL European
Feed (inc SuperBowl) at
40Mbit/s
27. Example of IT and Generic IP at
Scale (2)• Three nights of the One
Show on BBC1 delivered
using public internet (~5m
viewers)
• Reverse vision video on
same unit.
• Next step (in UK), long-
form programming
delivered over cellular?
28. Summary
• IT equipment is on-air in the live broadcast chain today
• The advent of IP-based production provides an opportunity
to use multifunction IT equipment and gain massive
flexibility.
• Standards need to reflect this
• The use of generic connectivity is increasing, and not just for
financial reasons.
• Going to be some very high-profile announcements in
coming months
Feel free to ask difficult questions at the end.
Tweeting is especially encouraged.
Plenty of IT in OTT.
Just because you write 1mbit/s OTT software, doesn’t make you qualified even to work with 10Mbps, let alone 100 or 1Gigabit.
Building a road-car is not the same as building an F1 car.
Hopefully something for everyone
SpaceX are vertically integrated (metal comes in, rockets come out). Parts and software built in-house.
Do things that commentators said was not possible.
Homage in a lot of places
Some manufacturers reflash fixed-function hardware on the fly.
In a few cases the boards do a little more than just convert physical to digital. Exception that proves the rule.
All of these things at the same time
Note, no GPUs. I’ll address the V-word later.
Could do trickery such as reuse of motion vectors
Not an SDI frontend to an analogue chipset like some boards.
Unthinkable a short time ago.
I’d like to dispel some myths
Pixels spanning packets makes things awkward
RFC written by software people so doesn’t have this constraint.
All of these pixel formats are specific to broadcast industry.
That’s not a percentage sign, that’s a times. A big sudoku puzzle.
Interop presentation, let’s look at technical reality
People understanding that upgrading live broadcast chains weekly is not something to be afraid of.
Just because PTP is a thing, doesn’t mean you should use it, or it will be available
Used to be financial, starting to become flexibility above costs.