SlideShare une entreprise Scribd logo
1  sur  56
Télécharger pour lire hors ligne
© 2016, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
November 28, 2016
High Performance Cinematic
Production in the Cloud
Richard Welsh, CEO, Sundog Media Toolkit
Konstantin Wilms, Principal Solutions Architect, AWS
MAE 304
What to Expect from this Session
Production on AWS
• Macro Perspective
• Building Blocks for Post-Production Pipelines
Sundog Media Toolkit
• Platform, Workflows, Scale, Production Techniques
• Billy Lynn’s Long Halftime Walk
• Challenges & Benefits
noun | film·mak·ing | -ˌmā-kiŋ
Filmmaking
The process of making a film incorporating five major stages:
Development, Pre-Production, Production, Post-Production, & Distribution
Benefit from massive
economies of scale
Adapt your media storage and
compute needs using AWS
Stop guessing capacity
Handle unpredictable & bursty
media needs programmatically
Trade capital expense for variable
expense
Pay for media you store and
process, as you go
Go global in minutes
Global availability instantly,
with no commit
Stop spending money on running &
maintaining data centers
Focus your resources on your media needs
Increase speed and agility
Shorten time to market,
test new approaches
Why AWS for Post Production?
Playout & Distribution
Visual Effects & Editing Analytics
DAM & Archive
Digital Supply Chain
Publishing
OTT
Acquisition
AWS Media Workloads
Media – ALL IN – to the Cloud
The New Normal
Media – ALL IN – to the Cloud
Where Do I Begin?
Buzzword?
Specific scope
Interoperable services
Allow for rapid innovation
De-Coupling
Micro & Composable Services
Performance Optimization
Storage
Render Farm
Transcode
Metadata
Pipeline and License
Manager
Graphics Artist
Workstations
• Content has gravity
• Network Bandwidth
• Hybrid/All-in Cloud
• I/O Performance
• Ability to burst at a very
short notice
• Cost?
• Performance
• Security
• License mobility/Elasticity
• Dependency Management
(hybrid scenario)
• Interactivity
• High-Performant Storage
• Hardware Support
Post-Production Components
All-In Cloud Enabled Domino Effect
PIPELINES
Mapping Studio Workflows to AWS & Partner Offerings
Ingest Storage Editing Processing
PUSH OR PULL
CONTENT & ARTIFACTS
ACCESS SOURCE FILES
LOCATED IN AMAZON S3
CREATE, EDIT, &
COLLABORATE IN THE CLOUD
SCALE OUT ON ELASTIC
CAPACITY FOR PROCESSING
Content production and post-production companies are leveraging AWS to accelerate and streamline
creative, editing, and rendering workloads with highly scalable cloud computing and storage.
Post-Production Pipeline
Ingest
PUSH OR PULL
CONTENT & ARTIFACTS
AWS Services
• Amazon S3 Transfer Acceleration: Fast data transfer to S3
• AWS Direct Connect: Dedicated network connectivity
• AWS Snowball: Petabyte-scale data transfer
• Amazon CloudFront: Global content delivery network
• AWS Storage Gateway: On-prem-to-cloud storage connectivity
• Elemental Cloud Appliances: On-prem video processing for ingest
Partners
Post-Production Pipeline
AWS Services
• Amazon S3: Highly scalable, durable object storage
• Amazon Glacier: Secure, durable, petabyte-scale data archival
• Amazon EFS: Simple, elastic, scalable file system
• Amazon Elastic Block Store: Persistent block storage
• AWS Marketplace: Partner solutions for storage
Partners
Storage
ACCESS SOURCE FILES
LOCATED IN S3
Post-Production Pipeline
AWS Services
• Amazon EC2: Scalable CPU/GPU computing capacity
• Amazon Workspaces: Virtual desktops in the cloud
• Amazon EFS: Simple, elastic, scalable file system
• Amazon EBS: Persistent block storage
• Amazon S3: Highly scalable, durable object storage
Partners
Editing
CREATE, EDIT, &
COLLABORATE IN THE CLOUD
Post-Production Pipeline
Processing
SCALE OUT ON ELASTIC
CAPACITY FOR PROCESSING
AWS Services
• Amazon EC2: Scalable CPU/GPU/Spot computing capacity
• Amazon EFS: Simple, elastic, scalable file system
• Amazon EBS: Persistent block storage
• Amazon S3: Highly scalable, durable object storage
• Elemental Cloud: On-demand, scalable video processing
Partners
Post-Production Pipeline
STORAGE
Content storage options on AWS for post-production pipelines
1 PB raw storage
800 TB usable storage
600 TB allocated storage
400 TB application data
Cloud Storage
Storage Capacity
Storage Services
Amazon EBS
Block storage for use
with Amazon EC2
Amazon S3
Massively scalable
storage & front-end
11 9’s of durability
IA for infrequent access
Internet scale
storage via API
AWS Storage Gateway
S3,
Amazon
Glacier
Integrates on-premises
IT and AWS storage
Amazon Glacier
$0.004/GB/month
11 9’s of durability
Multiple copies across
different DCs
Storage for archiving and
backup
EC2
EBS
Amazon EFS
Share File storage for
use with Amazon EC2
EC2
EFS
Massively scalable
Pay for what you use
High Performance
1000’s of hosts
Up to 16TB/volume
Up to 20K PIOPS
SSD-backed
Cold & Throughput
Encryption
BLOCK FILE OBJECT OBJECT
VPCVPCVPC
Asset Lifecycle
Long-Term
Archive
All tiers
accessible through
a single API
Oldest content trickles
down to Amazon Glacier
automatically to save cost
Amazon S3
S3 Infrequent Access
Amazon Glacier
LifeCyclePolicies
Near-Line
Storage
Frequently
Accessed
Shared Storage Shared Storage
On-prem Storage
AWS
Direct Connect
Storage Cache
Amazon S3
Lustre on EC2
Avere on EC2
EFS
AWS
Direct Connect
Hydrate workers
EC2 Spot
FXT on-prem
Access Models
Tightly vs. Loosely Coupled Storage
Post-Production tasks
NLE/Review & Approve/VFX
(Hybrid, on premises, or in Cloud)
Raw Assets
Secondary Copy
Amazon Glacier in another Region
Raw Assets
Primary Copy
Amazon Glacier
Finished (Golden Copy) Archive
Primary Copy
Amazon Glacier
AssetManagement(DAM/MAM)
(Hybrid,onpremises,orinCloud)
Active Archive
Amazon S3
(Standard)
Active Archive
Amazon S3
(Infrequent Access)
Secondary & Tertiary Copies
Amazon Glacier in another Region
A secondary offsite copy of the active archive – usually cold
Used for DR purposes
MediaSupplyChain
(Hybrid,onpremisesorincloud)
B2BDeliveryoverDXorInternet
Amazon Glacieringest
Affiliates/Third-Party Vendors
(Primary Copy) AWS Region 1 (Secondary & Tertiary Copies – optional)
AWS Region 2 or on-premises
Production Archive
deliver
Object Store
(Amazon S3)
Block Store
(Amazon EBS)
Shared FS
(Amazon EFS)
Performance
Per-Operation Latency Variable Lowest, Consistent Low, Consistent
Throughput Scale Multiple GBs per sec Single GBs per sec Multiple GBs per sec
Data
Availability/Durability
Across Multi-AZs Redundant Within an AZ Across Multi-AZs
Characteristics
Access 1-1000’s of Clients A Single EC2 instance 1-1000’s of Clients
Use Cases
Active Archive and
Lifecycle to Cold Archive
Media Processing
(Transcoding, Rendering,
QC, Farms) Depending on
Application Conformity
Databases
Boot Volumes
Media Processing
(Transcoding, Rendering,
QC, Farms) Singleton
Streaming
Data Analytics
Media Processing
(Transcoding, Rendering,
QC, Farms) Distributed
Content Management
Web Serving/Publishing
Workload Specific Storage
EDITING
Using AWS compute for content & media post-production
Editing in the Cloud
Amazon WorkSpaces
NVIDIA K520 GPU, 16GB RAM, 8 VCPUs
Heatmaps
H.264 vs. PCoIP
• PCoIP codecs were designed for interactive applications that are often static and are
dominated by computer-generated content
• Video codecs (like H.264) are designed for passive streaming of constantly changing, natural-
image camera-captured content
Designed for graphically intense cloud delivery
• The PCoIP protocol delivers
fundamentally better image
quality for text and 3D graphics
• The PCoIP protocol is optimized for
cloud delivery:
• Consumes 60-80% less bandwidth for
text and graphics
• Requires 50+% less server CPU,
leaving more cycles for the
applications
Cloud Editing Protocols
PROCESSING
Using AWS compute for content & media post-production
Reserved
Make a low, one-time
payment and receive
a significant discount
on the hourly charge
For committed
utilization
On-Demand
Pay for compute
capacity by the hour
with no long-term
commitments
For spiky workloads,
or to define needs
Spot
Bid for unused capacity
at a Spot Price -
fluctuates based on
supply & demand
For time-insensitive or
transient workloads
Consumption Models
Compute Intensive
Intel ES-2666 v3 (Haswell)
Optimized specifically for EC2
Memory Intensive
Lowest price point per GiB of RAM
GPUs
40K CUDA cores
192 GB of video memory
Enhanced Networking
Higher PPS, lower network jitter, low latency
I/O Intensive
SSD Storage, EBS Optimized
High Storage
24 x 2000 GiB per instance
Amazon S3
Amazon SQS Queues (orchestrated by Amazon SWF)
M4
M3
C4
C3 I3G2
R4
P2 D2
R3
X1
Task Optimization
Accelerated Computing
Rendering, Compositing, Encoding, Non-Linear Editing, VR
*12-core E5-2697-v2 Intel Xeon
Why GPUs?
• Ubiquitous – available at high scale to
application developers worldwide
• Enabling a high degree of parallelism – each
GPU has thousands of cores
• Consistent, well-documented set of APIs (CUDA,
OpenACC, OpenCL)
• Supported by a wide variety of ISVs and open
source frameworks
• Deploy as Spot for compute workloads
• Single K80 10x faster than CPU on applications*
CPU vs. GPU Rendering
Scheduling
• Multi-threaded, multi-process
• Select instance based on thread #s
• Or map across GPUs:
CUDA_0: Frame 1-25
CUDA_1: Frame 26-50
• Increase in Scheduler Complexity
Dependencies
• File loading - CPU/storage
• Package, meshes, textures - CPU
• 3D rendering - CPU/GPU
• Compositing - CPU
• Image export - CPU/storage
SUNDOG MEDIA TOOLKIT
An On-Demand Cloud-Based Post-Production Platform
Asset Management
Workflows
Collaborative Review
Playlist Mastering
3D Visualization
Services Platform
• Full Stack Public
• Hybrid
Storage Architecture
Tool Identity
Workflow Dependencies
Tool Handling / Reporting
Data I/O
Metadata Handling
Content Handling
Process
Core
Workflow Tool Shell
Encryption Points
Workflow Manager Worker (Primary)UI
Worker (Clone)
Worker (Clone)
Worker (Clone)
Worker (Clone)
Worker (Clone)
Worker (Clone)
Worker (Clone)
Worker (Clone)
Worker (Clone)
Worker (Clone)
Worker (Clone)
Worker (Primary)
Workflow Creation Workflow Analysis Worker Spawning
Availability Zone Spanning
(availability/outage)
Job Completion
Worker Flushing
Status/ProgressReporting
Worker (Clone)
Worker (Clone)
Worker (Clone)
Worker (Clone)
Worker (Clone)
Worker (Clone)
Primary zone (a)
ZonecZoneb
Scale Decisions:
Tool process
Dependencies (intra and inter-
tool)
Size
- bit depth
- resolution
- layers
Frame count
Elasticresourceexpansion
Workflow & Scale Decision Process
Iterate
Inspect
pixels
Vicinity (spatial)
Scene-Based
Object-Based
Including 3D
Colour (spectral)
Motion (temporal)
Analyse to sub-pixel level the statistical consistency
of that pixel with the expected value
Make a fine adjustment
Multi-Mode Analysis Super Pixel
Reconstruction
Spectral Reconstruction
Spectral Reconstruction
Temporal Reconstruction of Detail
Spatial Reconstruction
Spatial Reconstruction
Source
Blurred
Sharpened
TrueImage Processed
Original = Baseline
Blurred = PSNR high, SSIM low
Sharpened = PSNR low, SSIM high
TrueImage = PSNR high, SSIM high
Compression Efficiency
• 120 Frames Per Second
• 3D (240 real FPS)
• 4K (4096x2160 – not UHD “4K”)
• High Dynamic Range (16-bit float)
Uncompressed Data Rate
= 40 x average cinema
= 55 x average HD
Billy Lynn’s Long Halftime Walk
Billy Lynn’s Long Halftime Walk
Iterations
Datapointsperframe(4K)
Data points
per pixel
Trillion
(x1012)
10.1
21.9
28.0
56.1
Challenges
• Scaling to 120fps 3D 4K uncompressed data processing
• Breaking traditional workflows (i.e., traditional thinking)
• Staying Safe
Conclusion
Customer Feedback
“Sundog cloud based workflows have enhanced our ability to deliver
an improved product for our clients and are a massive step change in
how we go about our business.”
“TrueImage exceeded all expectations and is the primary reason that
the movie looks so good on the big screen … it actually had a powerful
effect on the editorial and story telling”
Conclusion
Find Us!
sundogtools.com
@sundogmt
Thank you!
Remember to complete your
evaluations!
Industry Pre-Day: Monday 11/28
• MAE303 - Discovery Channel's Broadcast Workflows and Channel
Origination on AWS
• MAE302 - Turner's Cloud Native Media Supply Chain for TNT, TBS, Adult
Swim, Cartoon Network, CNN
• MAE301 - Accelerating the Transition to Broadcast and OTT Infrastructure
in the Cloud
Regular Conference: Tuesday – Thursday 11/29 – 12/1
• ARC406 - Encoding Artifacts to Emmy Awards: Taking on Terabyte-Scale, 1-
Gbps, 4K Video Processing in the Cloud
• CMP312 - Powering the Next Generation of Virtual Reality with Verizon
• CTD305 - Media Delivery from the Cloud: Integrated AWS Solutions for
Premium Over the Top (OTT) Content
• STG302 - Deep Dive on Amazon Glacier
Related Sessions

Contenu connexe

Tendances

Tendances (20)

AWS re:Invent 2016: Workshop: Addressing Your Business Needs with AWS (ARC210)
AWS re:Invent 2016: Workshop: Addressing Your Business Needs with AWS (ARC210)AWS re:Invent 2016: Workshop: Addressing Your Business Needs with AWS (ARC210)
AWS re:Invent 2016: Workshop: Addressing Your Business Needs with AWS (ARC210)
 
AWS re:Invent 2016: Turner's cloud native media supply chain for TNT, TBS, Ad...
AWS re:Invent 2016: Turner's cloud native media supply chain for TNT, TBS, Ad...AWS re:Invent 2016: Turner's cloud native media supply chain for TNT, TBS, Ad...
AWS re:Invent 2016: Turner's cloud native media supply chain for TNT, TBS, Ad...
 
AWS re:Invent 2016: How Fulfillment by Amazon (FBA) and Scopely Improved Resu...
AWS re:Invent 2016: How Fulfillment by Amazon (FBA) and Scopely Improved Resu...AWS re:Invent 2016: How Fulfillment by Amazon (FBA) and Scopely Improved Resu...
AWS re:Invent 2016: How Fulfillment by Amazon (FBA) and Scopely Improved Resu...
 
Migrating Your Databases to AWS Deep Dive on Amazon RDS and AWS
Migrating Your Databases to AWS Deep Dive on Amazon RDS and AWSMigrating Your Databases to AWS Deep Dive on Amazon RDS and AWS
Migrating Your Databases to AWS Deep Dive on Amazon RDS and AWS
 
Getting Started with Managed Database Services on AWS - September 2016 Webina...
Getting Started with Managed Database Services on AWS - September 2016 Webina...Getting Started with Managed Database Services on AWS - September 2016 Webina...
Getting Started with Managed Database Services on AWS - September 2016 Webina...
 
Born in the Cloud; Build it Like a Startup
Born in the Cloud; Build it Like a StartupBorn in the Cloud; Build it Like a Startup
Born in the Cloud; Build it Like a Startup
 
SAP Workloads on the AWS Cloud - AWS Innovate Toronto
SAP Workloads on the AWS Cloud - AWS Innovate TorontoSAP Workloads on the AWS Cloud - AWS Innovate Toronto
SAP Workloads on the AWS Cloud - AWS Innovate Toronto
 
SRV409 Deep Dive on Microservices and Docker
SRV409 Deep Dive on Microservices and DockerSRV409 Deep Dive on Microservices and Docker
SRV409 Deep Dive on Microservices and Docker
 
The Best of re:invent 2016
The Best of re:invent 2016The Best of re:invent 2016
The Best of re:invent 2016
 
AWS re:Invent 2016: Continuous Compliance in the AWS Cloud for Regulated Life...
AWS re:Invent 2016: Continuous Compliance in the AWS Cloud for Regulated Life...AWS re:Invent 2016: Continuous Compliance in the AWS Cloud for Regulated Life...
AWS re:Invent 2016: Continuous Compliance in the AWS Cloud for Regulated Life...
 
AWS re:Invent 2016: Accelerating the Transition to Broadcast and OTT Infrastr...
AWS re:Invent 2016: Accelerating the Transition to Broadcast and OTT Infrastr...AWS re:Invent 2016: Accelerating the Transition to Broadcast and OTT Infrastr...
AWS re:Invent 2016: Accelerating the Transition to Broadcast and OTT Infrastr...
 
Database and Analytics on the AWS Cloud - AWS Innovate Toronto
Database and Analytics on the AWS Cloud - AWS Innovate TorontoDatabase and Analytics on the AWS Cloud - AWS Innovate Toronto
Database and Analytics on the AWS Cloud - AWS Innovate Toronto
 
AWS re:Invent 2016: Wild Rydes Takes Off – The Dawn of a New Unicorn (SVR309)
AWS re:Invent 2016: Wild Rydes Takes Off – The Dawn of a New Unicorn (SVR309)AWS re:Invent 2016: Wild Rydes Takes Off – The Dawn of a New Unicorn (SVR309)
AWS re:Invent 2016: Wild Rydes Takes Off – The Dawn of a New Unicorn (SVR309)
 
Welcome Keynote - AWS Summit Stockholm
Welcome Keynote - AWS Summit Stockholm Welcome Keynote - AWS Summit Stockholm
Welcome Keynote - AWS Summit Stockholm
 
Deep Dive on Amazon Relational Database Service
Deep Dive on Amazon Relational Database ServiceDeep Dive on Amazon Relational Database Service
Deep Dive on Amazon Relational Database Service
 
ENT201 A Tale of Two Pizzas: Accelerating Software Delivery with AWS Develope...
ENT201 A Tale of Two Pizzas: Accelerating Software Delivery with AWS Develope...ENT201 A Tale of Two Pizzas: Accelerating Software Delivery with AWS Develope...
ENT201 A Tale of Two Pizzas: Accelerating Software Delivery with AWS Develope...
 
BDA302 Deep Dive on Migrating Big Data Workloads to Amazon EMR
BDA302 Deep Dive on Migrating Big Data Workloads to Amazon EMRBDA302 Deep Dive on Migrating Big Data Workloads to Amazon EMR
BDA302 Deep Dive on Migrating Big Data Workloads to Amazon EMR
 
(ARC305) How J&J Manages AWS At Scale For Enterprise Workloads
(ARC305) How J&J Manages AWS At Scale For Enterprise Workloads(ARC305) How J&J Manages AWS At Scale For Enterprise Workloads
(ARC305) How J&J Manages AWS At Scale For Enterprise Workloads
 
The State of Serverless Computing | AWS Public Sector Summit 2017
The State of Serverless Computing | AWS Public Sector Summit 2017The State of Serverless Computing | AWS Public Sector Summit 2017
The State of Serverless Computing | AWS Public Sector Summit 2017
 
Selecting the Right AWS Database Solution - AWS 2017 Online Tech Talks
Selecting the Right AWS Database Solution - AWS 2017 Online Tech TalksSelecting the Right AWS Database Solution - AWS 2017 Online Tech Talks
Selecting the Right AWS Database Solution - AWS 2017 Online Tech Talks
 

En vedette

En vedette (20)

AWS re:Invent 2016: Journeys to the Cloud: Different Experiences in Video (CT...
AWS re:Invent 2016: Journeys to the Cloud: Different Experiences in Video (CT...AWS re:Invent 2016: Journeys to the Cloud: Different Experiences in Video (CT...
AWS re:Invent 2016: Journeys to the Cloud: Different Experiences in Video (CT...
 
Building Scalable Immersive Media Solutions on AWS
Building Scalable Immersive Media Solutions on AWSBuilding Scalable Immersive Media Solutions on AWS
Building Scalable Immersive Media Solutions on AWS
 
AWS re:Invent 2016: Discovery Channel's Broadcast Workflows and Channel Origi...
AWS re:Invent 2016: Discovery Channel's Broadcast Workflows and Channel Origi...AWS re:Invent 2016: Discovery Channel's Broadcast Workflows and Channel Origi...
AWS re:Invent 2016: Discovery Channel's Broadcast Workflows and Channel Origi...
 
AWS re:Invent 2016: Open-Source Resources (DCS201)
AWS re:Invent 2016: Open-Source Resources (DCS201)AWS re:Invent 2016: Open-Source Resources (DCS201)
AWS re:Invent 2016: Open-Source Resources (DCS201)
 
WebSocket Perspectives and Vision for the Future
WebSocket Perspectives and Vision for the FutureWebSocket Perspectives and Vision for the Future
WebSocket Perspectives and Vision for the Future
 
AWS Cloud for HPC and Big Data
AWS Cloud for HPC and Big DataAWS Cloud for HPC and Big Data
AWS Cloud for HPC and Big Data
 
Cloud TV playout for disaster recovery
Cloud TV playout for disaster recoveryCloud TV playout for disaster recovery
Cloud TV playout for disaster recovery
 
CUMULUS Cloud Broadcast Platform
CUMULUS Cloud Broadcast PlatformCUMULUS Cloud Broadcast Platform
CUMULUS Cloud Broadcast Platform
 
Monetizing public cloud
Monetizing public cloudMonetizing public cloud
Monetizing public cloud
 
Next Generation Audience Measurement at Spectrum Reach
Next Generation Audience Measurement at Spectrum ReachNext Generation Audience Measurement at Spectrum Reach
Next Generation Audience Measurement at Spectrum Reach
 
Igniting Audience Measurement at Time Warner Cable
Igniting Audience Measurement at Time Warner CableIgniting Audience Measurement at Time Warner Cable
Igniting Audience Measurement at Time Warner Cable
 
Alert Logic
Alert LogicAlert Logic
Alert Logic
 
Software Ecosystem Evolution. It's complex!
Software Ecosystem Evolution. It's complex!Software Ecosystem Evolution. It's complex!
Software Ecosystem Evolution. It's complex!
 
QuESo: a Quality Model for Open Source Software Ecosystems
QuESo: a Quality Model for Open Source Software EcosystemsQuESo: a Quality Model for Open Source Software Ecosystems
QuESo: a Quality Model for Open Source Software Ecosystems
 
AWS re:Invent 2016: Using AWS to Meet Requirements for Education, Healthcare ...
AWS re:Invent 2016: Using AWS to Meet Requirements for Education, Healthcare ...AWS re:Invent 2016: Using AWS to Meet Requirements for Education, Healthcare ...
AWS re:Invent 2016: Using AWS to Meet Requirements for Education, Healthcare ...
 
Intro to High Performance Computing in the AWS Cloud
Intro to High Performance Computing in the AWS CloudIntro to High Performance Computing in the AWS Cloud
Intro to High Performance Computing in the AWS Cloud
 
음성인식 기반 홈IoT 제품서비스 동향 및 시사점 스마트홈산업협회 (순천향대학교 김학용 교수)
음성인식 기반 홈IoT 제품서비스 동향 및 시사점   스마트홈산업협회 (순천향대학교 김학용 교수)음성인식 기반 홈IoT 제품서비스 동향 및 시사점   스마트홈산업협회 (순천향대학교 김학용 교수)
음성인식 기반 홈IoT 제품서비스 동향 및 시사점 스마트홈산업협회 (순천향대학교 김학용 교수)
 
Microsoft vision & strategy keynote for partners
Microsoft vision & strategy keynote for partnersMicrosoft vision & strategy keynote for partners
Microsoft vision & strategy keynote for partners
 
Cost Effective Rendering in the Cloud with Spot Instances
Cost Effective Rendering in the Cloud with Spot InstancesCost Effective Rendering in the Cloud with Spot Instances
Cost Effective Rendering in the Cloud with Spot Instances
 
AWS Enabling the Startup Ecosystem - AWS India Summit 2012
AWS Enabling the Startup Ecosystem - AWS India Summit 2012AWS Enabling the Startup Ecosystem - AWS India Summit 2012
AWS Enabling the Startup Ecosystem - AWS India Summit 2012
 

Similaire à AWS re:Invent 2016: High Performance Cinematic Production in the Cloud (MAE304)

Lunch and Learn - Store and Move your Data To & From the AWS Cloud, Markku Le...
Lunch and Learn - Store and Move your Data To & From the AWS Cloud, Markku Le...Lunch and Learn - Store and Move your Data To & From the AWS Cloud, Markku Le...
Lunch and Learn - Store and Move your Data To & From the AWS Cloud, Markku Le...
Amazon Web Services
 

Similaire à AWS re:Invent 2016: High Performance Cinematic Production in the Cloud (MAE304) (20)

AWS Services for Content Production
AWS Services for Content ProductionAWS Services for Content Production
AWS Services for Content Production
 
Moving Your Media Supply Chain to the AWS Cloud
Moving Your Media Supply Chain to the AWS CloudMoving Your Media Supply Chain to the AWS Cloud
Moving Your Media Supply Chain to the AWS Cloud
 
Lunch and Learn - Store and Move your Data To & From the AWS Cloud, Markku Le...
Lunch and Learn - Store and Move your Data To & From the AWS Cloud, Markku Le...Lunch and Learn - Store and Move your Data To & From the AWS Cloud, Markku Le...
Lunch and Learn - Store and Move your Data To & From the AWS Cloud, Markku Le...
 
AWS Webinar 201 - Backup, Archive and Disaster Recovery
AWS Webinar 201 - Backup, Archive and Disaster RecoveryAWS Webinar 201 - Backup, Archive and Disaster Recovery
AWS Webinar 201 - Backup, Archive and Disaster Recovery
 
4K Media Workflows on AWS
4K Media Workflows on AWS4K Media Workflows on AWS
4K Media Workflows on AWS
 
Backup and archiving in the aws cloud
Backup and archiving in the aws cloudBackup and archiving in the aws cloud
Backup and archiving in the aws cloud
 
Understanding AWS Storage Options (STG101) | AWS re:Invent 2013
Understanding AWS Storage Options (STG101) | AWS re:Invent 2013Understanding AWS Storage Options (STG101) | AWS re:Invent 2013
Understanding AWS Storage Options (STG101) | AWS re:Invent 2013
 
Transitioning to the Next Generation Hybrid Cloud Operating Model
Transitioning to the Next Generation Hybrid Cloud Operating ModelTransitioning to the Next Generation Hybrid Cloud Operating Model
Transitioning to the Next Generation Hybrid Cloud Operating Model
 
AWS Webcast - How to Migrate On-premise NAS Storage to Cloud NAS Storage
AWS Webcast - How to Migrate On-premise NAS Storage to Cloud NAS StorageAWS Webcast - How to Migrate On-premise NAS Storage to Cloud NAS Storage
AWS Webcast - How to Migrate On-premise NAS Storage to Cloud NAS Storage
 
Create cloud service on AWS
Create cloud service on AWSCreate cloud service on AWS
Create cloud service on AWS
 
AWS Summit 2013 | Auckland - Scalable Media Processing on the Cloud
AWS Summit 2013 | Auckland - Scalable Media Processing on the CloudAWS Summit 2013 | Auckland - Scalable Media Processing on the Cloud
AWS Summit 2013 | Auckland - Scalable Media Processing on the Cloud
 
Scalable Media Workflows in the Cloud
Scalable Media Workflows in the CloudScalable Media Workflows in the Cloud
Scalable Media Workflows in the Cloud
 
ENT306 Migrating large Scale Data Sets to the Cloud
ENT306 Migrating large Scale Data Sets to the CloudENT306 Migrating large Scale Data Sets to the Cloud
ENT306 Migrating large Scale Data Sets to the Cloud
 
Creative content storage in the AWS Cloud
Creative content storage in the AWS CloudCreative content storage in the AWS Cloud
Creative content storage in the AWS Cloud
 
AWS May Webinar Series - Getting Started: Storage with Amazon S3 and Amazon G...
AWS May Webinar Series - Getting Started: Storage with Amazon S3 and Amazon G...AWS May Webinar Series - Getting Started: Storage with Amazon S3 and Amazon G...
AWS May Webinar Series - Getting Started: Storage with Amazon S3 and Amazon G...
 
AWS Services for Content Production
AWS Services for Content ProductionAWS Services for Content Production
AWS Services for Content Production
 
Harness the Power of Hybrid Cloud with AWS and Avere
Harness the Power of Hybrid Cloud with AWS and AvereHarness the Power of Hybrid Cloud with AWS and Avere
Harness the Power of Hybrid Cloud with AWS and Avere
 
Scaling the Platform for Your Startup - Startup Talks June 2015
Scaling the Platform for Your Startup - Startup Talks June 2015Scaling the Platform for Your Startup - Startup Talks June 2015
Scaling the Platform for Your Startup - Startup Talks June 2015
 
AWS Sydney Summit 2013 - Understanding your AWS Storage Options
AWS Sydney Summit 2013 - Understanding your AWS Storage OptionsAWS Sydney Summit 2013 - Understanding your AWS Storage Options
AWS Sydney Summit 2013 - Understanding your AWS Storage Options
 
Common Workloads on the AWS Cloud
Common Workloads on the AWS CloudCommon Workloads on the AWS Cloud
Common Workloads on the AWS Cloud
 

Plus de Amazon Web Services

Tools for building your MVP on AWS
Tools for building your MVP on AWSTools for building your MVP on AWS
Tools for building your MVP on AWS
Amazon Web Services
 
How to Build a Winning Pitch Deck
How to Build a Winning Pitch DeckHow to Build a Winning Pitch Deck
How to Build a Winning Pitch Deck
Amazon Web Services
 
Building a web application without servers
Building a web application without serversBuilding a web application without servers
Building a web application without servers
Amazon Web Services
 
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
Amazon Web Services
 

Plus de Amazon Web Services (20)

Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
 
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
 
Esegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS FargateEsegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS Fargate
 
Costruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWSCostruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWS
 
Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot
 
Open banking as a service
Open banking as a serviceOpen banking as a service
Open banking as a service
 
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
 
OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...
 
Microsoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows WorkloadsMicrosoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows Workloads
 
Computer Vision con AWS
Computer Vision con AWSComputer Vision con AWS
Computer Vision con AWS
 
Database Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatareDatabase Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatare
 
Crea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJSCrea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJS
 
API moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e webAPI moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e web
 
Database Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatareDatabase Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatare
 
Tools for building your MVP on AWS
Tools for building your MVP on AWSTools for building your MVP on AWS
Tools for building your MVP on AWS
 
How to Build a Winning Pitch Deck
How to Build a Winning Pitch DeckHow to Build a Winning Pitch Deck
How to Build a Winning Pitch Deck
 
Building a web application without servers
Building a web application without serversBuilding a web application without servers
Building a web application without servers
 
Fundraising Essentials
Fundraising EssentialsFundraising Essentials
Fundraising Essentials
 
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
 
Introduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container ServiceIntroduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container Service
 

Dernier

Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire business
panagenda
 

Dernier (20)

A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost SavingRepurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
 
MS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectorsMS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectors
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdf
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
 
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024
 
Ransomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdfRansomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdf
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CV
 
Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire business
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Script
 
Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 

AWS re:Invent 2016: High Performance Cinematic Production in the Cloud (MAE304)

  • 1. © 2016, Amazon Web Services, Inc. or its Affiliates. All rights reserved. November 28, 2016 High Performance Cinematic Production in the Cloud Richard Welsh, CEO, Sundog Media Toolkit Konstantin Wilms, Principal Solutions Architect, AWS MAE 304
  • 2. What to Expect from this Session Production on AWS • Macro Perspective • Building Blocks for Post-Production Pipelines Sundog Media Toolkit • Platform, Workflows, Scale, Production Techniques • Billy Lynn’s Long Halftime Walk • Challenges & Benefits
  • 3. noun | film·mak·ing | -ˌmā-kiŋ Filmmaking The process of making a film incorporating five major stages: Development, Pre-Production, Production, Post-Production, & Distribution
  • 4. Benefit from massive economies of scale Adapt your media storage and compute needs using AWS Stop guessing capacity Handle unpredictable & bursty media needs programmatically Trade capital expense for variable expense Pay for media you store and process, as you go Go global in minutes Global availability instantly, with no commit Stop spending money on running & maintaining data centers Focus your resources on your media needs Increase speed and agility Shorten time to market, test new approaches Why AWS for Post Production?
  • 5. Playout & Distribution Visual Effects & Editing Analytics DAM & Archive Digital Supply Chain Publishing OTT Acquisition AWS Media Workloads
  • 6. Media – ALL IN – to the Cloud The New Normal
  • 7. Media – ALL IN – to the Cloud Where Do I Begin? Buzzword? Specific scope Interoperable services Allow for rapid innovation De-Coupling Micro & Composable Services Performance Optimization
  • 8. Storage Render Farm Transcode Metadata Pipeline and License Manager Graphics Artist Workstations • Content has gravity • Network Bandwidth • Hybrid/All-in Cloud • I/O Performance • Ability to burst at a very short notice • Cost? • Performance • Security • License mobility/Elasticity • Dependency Management (hybrid scenario) • Interactivity • High-Performant Storage • Hardware Support Post-Production Components All-In Cloud Enabled Domino Effect
  • 9. PIPELINES Mapping Studio Workflows to AWS & Partner Offerings
  • 10. Ingest Storage Editing Processing PUSH OR PULL CONTENT & ARTIFACTS ACCESS SOURCE FILES LOCATED IN AMAZON S3 CREATE, EDIT, & COLLABORATE IN THE CLOUD SCALE OUT ON ELASTIC CAPACITY FOR PROCESSING Content production and post-production companies are leveraging AWS to accelerate and streamline creative, editing, and rendering workloads with highly scalable cloud computing and storage. Post-Production Pipeline
  • 11. Ingest PUSH OR PULL CONTENT & ARTIFACTS AWS Services • Amazon S3 Transfer Acceleration: Fast data transfer to S3 • AWS Direct Connect: Dedicated network connectivity • AWS Snowball: Petabyte-scale data transfer • Amazon CloudFront: Global content delivery network • AWS Storage Gateway: On-prem-to-cloud storage connectivity • Elemental Cloud Appliances: On-prem video processing for ingest Partners Post-Production Pipeline
  • 12. AWS Services • Amazon S3: Highly scalable, durable object storage • Amazon Glacier: Secure, durable, petabyte-scale data archival • Amazon EFS: Simple, elastic, scalable file system • Amazon Elastic Block Store: Persistent block storage • AWS Marketplace: Partner solutions for storage Partners Storage ACCESS SOURCE FILES LOCATED IN S3 Post-Production Pipeline
  • 13. AWS Services • Amazon EC2: Scalable CPU/GPU computing capacity • Amazon Workspaces: Virtual desktops in the cloud • Amazon EFS: Simple, elastic, scalable file system • Amazon EBS: Persistent block storage • Amazon S3: Highly scalable, durable object storage Partners Editing CREATE, EDIT, & COLLABORATE IN THE CLOUD Post-Production Pipeline
  • 14. Processing SCALE OUT ON ELASTIC CAPACITY FOR PROCESSING AWS Services • Amazon EC2: Scalable CPU/GPU/Spot computing capacity • Amazon EFS: Simple, elastic, scalable file system • Amazon EBS: Persistent block storage • Amazon S3: Highly scalable, durable object storage • Elemental Cloud: On-demand, scalable video processing Partners Post-Production Pipeline
  • 15. STORAGE Content storage options on AWS for post-production pipelines
  • 16. 1 PB raw storage 800 TB usable storage 600 TB allocated storage 400 TB application data Cloud Storage Storage Capacity
  • 17. Storage Services Amazon EBS Block storage for use with Amazon EC2 Amazon S3 Massively scalable storage & front-end 11 9’s of durability IA for infrequent access Internet scale storage via API AWS Storage Gateway S3, Amazon Glacier Integrates on-premises IT and AWS storage Amazon Glacier $0.004/GB/month 11 9’s of durability Multiple copies across different DCs Storage for archiving and backup EC2 EBS Amazon EFS Share File storage for use with Amazon EC2 EC2 EFS Massively scalable Pay for what you use High Performance 1000’s of hosts Up to 16TB/volume Up to 20K PIOPS SSD-backed Cold & Throughput Encryption BLOCK FILE OBJECT OBJECT VPCVPCVPC
  • 18. Asset Lifecycle Long-Term Archive All tiers accessible through a single API Oldest content trickles down to Amazon Glacier automatically to save cost Amazon S3 S3 Infrequent Access Amazon Glacier LifeCyclePolicies Near-Line Storage Frequently Accessed
  • 19. Shared Storage Shared Storage On-prem Storage AWS Direct Connect Storage Cache Amazon S3 Lustre on EC2 Avere on EC2 EFS AWS Direct Connect Hydrate workers EC2 Spot FXT on-prem Access Models Tightly vs. Loosely Coupled Storage
  • 20. Post-Production tasks NLE/Review & Approve/VFX (Hybrid, on premises, or in Cloud) Raw Assets Secondary Copy Amazon Glacier in another Region Raw Assets Primary Copy Amazon Glacier Finished (Golden Copy) Archive Primary Copy Amazon Glacier AssetManagement(DAM/MAM) (Hybrid,onpremises,orinCloud) Active Archive Amazon S3 (Standard) Active Archive Amazon S3 (Infrequent Access) Secondary & Tertiary Copies Amazon Glacier in another Region A secondary offsite copy of the active archive – usually cold Used for DR purposes MediaSupplyChain (Hybrid,onpremisesorincloud) B2BDeliveryoverDXorInternet Amazon Glacieringest Affiliates/Third-Party Vendors (Primary Copy) AWS Region 1 (Secondary & Tertiary Copies – optional) AWS Region 2 or on-premises Production Archive deliver
  • 21. Object Store (Amazon S3) Block Store (Amazon EBS) Shared FS (Amazon EFS) Performance Per-Operation Latency Variable Lowest, Consistent Low, Consistent Throughput Scale Multiple GBs per sec Single GBs per sec Multiple GBs per sec Data Availability/Durability Across Multi-AZs Redundant Within an AZ Across Multi-AZs Characteristics Access 1-1000’s of Clients A Single EC2 instance 1-1000’s of Clients Use Cases Active Archive and Lifecycle to Cold Archive Media Processing (Transcoding, Rendering, QC, Farms) Depending on Application Conformity Databases Boot Volumes Media Processing (Transcoding, Rendering, QC, Farms) Singleton Streaming Data Analytics Media Processing (Transcoding, Rendering, QC, Farms) Distributed Content Management Web Serving/Publishing Workload Specific Storage
  • 22. EDITING Using AWS compute for content & media post-production
  • 23. Editing in the Cloud Amazon WorkSpaces NVIDIA K520 GPU, 16GB RAM, 8 VCPUs Heatmaps H.264 vs. PCoIP
  • 24. • PCoIP codecs were designed for interactive applications that are often static and are dominated by computer-generated content • Video codecs (like H.264) are designed for passive streaming of constantly changing, natural- image camera-captured content Designed for graphically intense cloud delivery • The PCoIP protocol delivers fundamentally better image quality for text and 3D graphics • The PCoIP protocol is optimized for cloud delivery: • Consumes 60-80% less bandwidth for text and graphics • Requires 50+% less server CPU, leaving more cycles for the applications Cloud Editing Protocols
  • 25. PROCESSING Using AWS compute for content & media post-production
  • 26. Reserved Make a low, one-time payment and receive a significant discount on the hourly charge For committed utilization On-Demand Pay for compute capacity by the hour with no long-term commitments For spiky workloads, or to define needs Spot Bid for unused capacity at a Spot Price - fluctuates based on supply & demand For time-insensitive or transient workloads Consumption Models
  • 27. Compute Intensive Intel ES-2666 v3 (Haswell) Optimized specifically for EC2 Memory Intensive Lowest price point per GiB of RAM GPUs 40K CUDA cores 192 GB of video memory Enhanced Networking Higher PPS, lower network jitter, low latency I/O Intensive SSD Storage, EBS Optimized High Storage 24 x 2000 GiB per instance Amazon S3 Amazon SQS Queues (orchestrated by Amazon SWF) M4 M3 C4 C3 I3G2 R4 P2 D2 R3 X1 Task Optimization
  • 28. Accelerated Computing Rendering, Compositing, Encoding, Non-Linear Editing, VR *12-core E5-2697-v2 Intel Xeon Why GPUs? • Ubiquitous – available at high scale to application developers worldwide • Enabling a high degree of parallelism – each GPU has thousands of cores • Consistent, well-documented set of APIs (CUDA, OpenACC, OpenCL) • Supported by a wide variety of ISVs and open source frameworks • Deploy as Spot for compute workloads • Single K80 10x faster than CPU on applications*
  • 29. CPU vs. GPU Rendering Scheduling • Multi-threaded, multi-process • Select instance based on thread #s • Or map across GPUs: CUDA_0: Frame 1-25 CUDA_1: Frame 26-50 • Increase in Scheduler Complexity Dependencies • File loading - CPU/storage • Package, meshes, textures - CPU • 3D rendering - CPU/GPU • Compositing - CPU • Image export - CPU/storage
  • 30. SUNDOG MEDIA TOOLKIT An On-Demand Cloud-Based Post-Production Platform
  • 31. Asset Management Workflows Collaborative Review Playlist Mastering 3D Visualization Services Platform
  • 32. • Full Stack Public • Hybrid Storage Architecture
  • 33. Tool Identity Workflow Dependencies Tool Handling / Reporting Data I/O Metadata Handling Content Handling Process Core Workflow Tool Shell
  • 35. Workflow Manager Worker (Primary)UI Worker (Clone) Worker (Clone) Worker (Clone) Worker (Clone) Worker (Clone) Worker (Clone) Worker (Clone) Worker (Clone) Worker (Clone) Worker (Clone) Worker (Clone) Worker (Primary) Workflow Creation Workflow Analysis Worker Spawning Availability Zone Spanning (availability/outage) Job Completion Worker Flushing Status/ProgressReporting Worker (Clone) Worker (Clone) Worker (Clone) Worker (Clone) Worker (Clone) Worker (Clone) Primary zone (a) ZonecZoneb Scale Decisions: Tool process Dependencies (intra and inter- tool) Size - bit depth - resolution - layers Frame count Elasticresourceexpansion Workflow & Scale Decision Process
  • 36. Iterate Inspect pixels Vicinity (spatial) Scene-Based Object-Based Including 3D Colour (spectral) Motion (temporal) Analyse to sub-pixel level the statistical consistency of that pixel with the expected value Make a fine adjustment Multi-Mode Analysis Super Pixel Reconstruction
  • 46. Original = Baseline Blurred = PSNR high, SSIM low Sharpened = PSNR low, SSIM high TrueImage = PSNR high, SSIM high Compression Efficiency
  • 47. • 120 Frames Per Second • 3D (240 real FPS) • 4K (4096x2160 – not UHD “4K”) • High Dynamic Range (16-bit float) Uncompressed Data Rate = 40 x average cinema = 55 x average HD Billy Lynn’s Long Halftime Walk
  • 48. Billy Lynn’s Long Halftime Walk
  • 49.
  • 51. Challenges • Scaling to 120fps 3D 4K uncompressed data processing • Breaking traditional workflows (i.e., traditional thinking) • Staying Safe Conclusion
  • 52. Customer Feedback “Sundog cloud based workflows have enhanced our ability to deliver an improved product for our clients and are a massive step change in how we go about our business.” “TrueImage exceeded all expectations and is the primary reason that the movie looks so good on the big screen … it actually had a powerful effect on the editorial and story telling” Conclusion
  • 55. Remember to complete your evaluations!
  • 56. Industry Pre-Day: Monday 11/28 • MAE303 - Discovery Channel's Broadcast Workflows and Channel Origination on AWS • MAE302 - Turner's Cloud Native Media Supply Chain for TNT, TBS, Adult Swim, Cartoon Network, CNN • MAE301 - Accelerating the Transition to Broadcast and OTT Infrastructure in the Cloud Regular Conference: Tuesday – Thursday 11/29 – 12/1 • ARC406 - Encoding Artifacts to Emmy Awards: Taking on Terabyte-Scale, 1- Gbps, 4K Video Processing in the Cloud • CMP312 - Powering the Next Generation of Virtual Reality with Verizon • CTD305 - Media Delivery from the Cloud: Integrated AWS Solutions for Premium Over the Top (OTT) Content • STG302 - Deep Dive on Amazon Glacier Related Sessions