4k media workflows on aws
TRANSCRIPT
4K Media Workflows on AWS
Usman ShakeelPrincipal Solutions Architect Lead (M&E)
Amazon Web Services
©2015, Amazon Web Services, Inc. or its affiliates. All rights reserved
Where is the problem?
More Bandwidth$$$$$
More PowerfulCompute $$$$$
Way more Storage$$$$$
Some Progress(ABR, HEVC, VP10)
A Concept - The Content LakeInspired from Data Lake (Coined by James Dixon in 2010)
A single store of all of digital content that you create and acquire in any form or factor•Don’t assume any resolutions/formats (for now or future)
•It is up to the consumer (application consuming the content) to use the appropriate infrastructure for processing
Amazon Simple Storage Service (S3) : the Content Lake
• Durable, cost-effective and fast
• Highly scalable front-end – Multi-part uploads (parallel writes)
– Range-gets (parallel reads)
• Use Amazon S3 with on-premises storage in a hybrid model
• No Storage capacity Upper-bounds
• Secure
Hydrating the Content Lake
Amazon S3
AWS Import/ ExportAmazon S3
(multi-part Upload)
AWS Storage Gateway
Direct Connect
N x 1G | 10G
Massively Scalable Front-end
Edge Locations
Availability Zone
Region
Dallas (2)
St.Louis
Miami
JacksonvilleLos Angeles (2)
Seattle
Ashburn (3)
Newark
New York (3)
Dublin
London (2)
Amsterdam (2)
Stockholm
Frankfurt (2)Paris (2)
Singapore(2)
Hong Kong (2)
Tokyo (2)
Sao Paulo
South Bend
San JosePalo AltoHayward
OsakaMilan
Sydney
MadridSeoul
Mumbai
Chennai
Regional Lakes …
Content Life-Cycle
Amazon Glacier (Life Cycle Policies)
Amazon S3
Near-term Storage
Long Term Archive
Both tiers accessible through a single API
Oldest content trickles down to glacier automatically to save cost
Amazon S3
Amazon S3 (range-gets)
Direct Connect
N x 1G | 10G
Massively Scalable S3 Front-end
EBS
Instance Store
cMassively Scalable Compute on AWS Cloud
On-Prem Apps
Consuming the Content Lake
Compute IntensiveIntel ES-2666 v3 (Haswell) optimized specifically for EC2
Memory IntensiveLowest price point per GiB of RAM
GPUs 1,536 CUDA cores4GB of video memory
Enhanced NetworkingHigher PPS, Lower network jitter, low latency
IO IntensiveSSD Storage, EBS Optimized
High Storage24 x 2000 GiB per instance
Massively Scalable Compute Layer
Amazon S3
c4g2
m3
HEVC Encode
GPU Transcode
HDi2
Package
Queues (orchestrated by Amazon SWF)
Content Lake
Task Specific
• Spot Instances(no charge if spot is terminated)
• Reserved Instances
• Pay for only what you use
• Buy for your need– Short term/Long term
– Real-time/Batch workloads
• AWS Marketplace – Pay for what licenses you use
Cost Effective Compute
VS. $2.80/hr
Launch a CloudFormation stack
with all the infrastructure
resources for a specific project
Autoscale the stack as
appropriate
AMI
CloudFormation
TemplateCloudFormation
Terminate
Template
Automated Infrastructure Layers based on Project Scope
AWS Ecosystem (License included in hourly* pricing)
INGEST STORE MANAGE SECUREPROCESS
CREATEMONETIZE
INTEGRATEDELIVER
AWS Import/ Export
S3 multi-part API
For Source, Renditions, Metadata Sidecar Files
Ingest
AWS ElasticBeanstalk
DAMWeb Service
CloudFront Distribution
Amazon DynamoDB
Amazon CloudSearch
DAM Catalog
Rendition Workers
EC2 ASG
Amazon SNS Topic
Amazon SQS QueueRendition Jobs
Amazon SQS QueueMetadata Processing Jobs
S3 Notification
Elastic TranscoderProxy / Thumbnail
Generation
Processing
Metadata Workers
EC2 ASG
A Simple DAM at AWS
AWS Import/ Export
S3 multi-part API
For Source, Renditions, Metadata Sidecar Files
Ingest
AWS ElasticBeanstalk
DAMWeb Service
CloudFront Distribution
Amazon DynamoDB
Amazon CloudSearch
DAM Catalog
Rendition Workers
EC2 ASG
Amazon SNS Topic
Amazon SQS QueueRendition Jobs
Amazon SQS QueueMetadata Processing Jobs
S3 Notification
Elastic TranscoderProxy / Thumbnail
Generation
Processing
Metadata Workers
EC2 ASG
A Simple DAM at AWS (Cost)
$
$
©2015, Amazon Web Services, Inc. or its affiliates. All rights reserved.
AWS Import/ Export
S3 multi-part API
For Source, Renditions, Metadata Sidecar Files
Ingest
AWS ElasticBeanstalk
DAMWeb Service
CloudFront Distribution
Amazon DynamoDB
Amazon CloudSearch
DAM Catalog
Rendition Workers
EC2 ASG
Amazon SNS Topic
Amazon SQS QueueRendition Jobs
Amazon SQS QueueMetadata Processing Jobs
S3 Notification
Elastic TranscoderProxy / Thumbnail
Generation
Processing
Metadata Workers
EC2 ASG
A Simple DAM at AWS (Using Spot)
0.$
0.$
AWS Import/ Export
S3 multi-part API
For Source, Renditions, Metadata Sidecar Files
Ingest
AWS Elastic
Beanstalk
DAM
Web Service
CloudFront
Distribution
S3 Notification
Elastic Transcoder
Proxy / Thumbnail
Generation
Processing
Lambda
Metadata
Extract & Persist
Amazon
DynamoDB
Amazon
CloudSearch
DAM Catalog
Server-less Applications using AWS Lambda
0.0$
yeah more science…
Remote application visualization for production tools, VFX modeling
Content stays in the Cloud
Powerful workstations in the cloud when needed
Amazon S3
GPU
Content Lake
Amazon Appstream
RemoteDumb Terminal
Amazon EC2
The Content Lake on AWS
Process
User Delivery/ConsumptionVFX/Production
On-Prem Apps
Archive
Content LakePartner/Affiliate/Service Provider
The Content Lake on AWS
EBS
Instance Store
Amazon Glacier (Life Cycle Policies)
Amazon S3
Amazon EBS/EC2 Instance Store
Process
User Delivery/ConsumptionVFX/Production
On-Prem Apps
Archive
Partner/Affiliate/Service Provider
Content Access Transfer
The Content Lake on AWS
EBS
Instance Store
Amazon Glacier (Life Cycle Policies) Amazon EBS/EC2 Instance Store
Process
Partner/Affiliate/Service Provider
User Delivery/ConsumptionVFX/Production
On-Prem Apps
Archive
Amazon Glacier (Life Cycle Policies) Amazon EBS/EC2 Instance Store
c
c
Direct Connect
Content Access Transfer
Disposable Infrastructure
AutoscalingWorkload specific
Amazon S3