Substantial Savings For LIVE Streaming
iSIZE Precoding software solution can save $176 per hour for every 5,000 viewers
iSIZE specializes in deep learning for video delivery and has developed a deep perceptual ‘precoder’ – a software solution that uses AI-trained to “see with the human eye” in order to optimize visual quality in order to save video bitrate during encoding. Its flagship product, BitSave, available both as a SaaS platform at bitsave.tech and for on-premise use, reduces encoding bitrates by up to 40% without compromising perceptual video quality. The technology is deployed as an add-on feature to conventional video encoding pipelines (AVC, HEVC and AV1), without requiring any changes in the streaming process or the client devices. This results in substantial bandwidth, energy and cost savings for VoD and live streaming services, broadcasters and the end consumers. You can see a presentation about our work in the AOMedia Research Forum here and read one of our first preprints here.
iSIZE is the first company to offer bolt-on and backward-compatible precoding solutions that allows video content to be sent and received at far lower transfer rates without any loss of quality. With video content predicted to make up 80% of the internet by 2022, it has never been more important to reduce the data, financial and energy costs of streaming.
More information is available from www.isize.co
The Saving Calculations For Live Streaming
We took AWS Live Streaming Cost Calculator and applied costs associated with iSIZE solution to calculate total savings that our customers that are streaming live channels can expect by implementing iSIZE BitSave solution
Assumptions:
Resolution: 1080p 30fps
Number of viewers: 5,000
Number of ABR Ladders: 9 (Apple-recommended TN2224 HLS parameters)
Let’s see how we can save $176 per hour for every 5,000 viewers.
Of the stated answers that are required, for this example, we’ll assume the requirement is to target a range of devices, including smartphones, tablets, computers and connected TVs. The contribution video stream to AWS will be a 1080p HD feed encoded at a bitrate of 15 Mbps using the AVC codec. For ABR (Adaptive Bitrate) encoding, we will use the Apple-recommended TN2224 HLS parameters, which are available as a preset within AWS Elemental MediaLive. This produces the following outputs:
Output 1 (_416x234_200k) |
Output 2 (_480x270_400k) |
Output 3 (_640x360_800k) |
Output 4 (_768x432_1200k) |
Output 5 (_960x540_2000k) |
Output 6 (_1280x720_3300k) |
Output 7 (_1280x720_5000k) |
Output 8 (_1280x720_6500k) |
Output 9 (_1920x1080_8000k) |
Since iSize precoder produces same quality videos in lower bitrates we will Precode the video first. Using CPU only iSize precoder requires 48 core instance for 1080p 30fps input so we will go for a c5d.12xlarge instance in the US East (N. Virginia) region with $2.304 per hour On-Demand pricing. The standard iSize precoder license in Amazon marketplace is $9/hour for a 48 core instance. Note that you only need to precode the video once irrespective of the number of ladders you plan to create later.
INPUT/OUTPUT | COST PER HOUR |
iSize Precoder | $9 |
c5d.12xlarge | $ 2.304 |
Total per hour | $11.31 |
This hourly cost is stable for this configuration
Now if we plan to encode with 25% bitrate saving over the standard rates then this produces the following outputs:
Output 1 (_416x234_150k) |
Output 2 (_480x270_300k) |
Output 3 (_640x360_600k) |
Output 4 (_768x432_900k) |
Output 5 (_960x540_1500k) |
Output 6 (_1280x720_2475) |
Output 7 (_1280x720_3750k) |
Output 8 (_1280x720_4875k) |
Output 9 (_1920x1080_6000k) |
We will use MediaLive’s built-in regional redundancy for a Standard channel, and will not run any addition channels to increase that level of redundancy. With these decisions made, we can start to work out some costs.
The hourly cost for running a MediaLive standard channel with this configuration in the US East (N. Virginia) region is $4.872 per hour with On-Demand pricing.
INPUT/OUTPUT | COST PER HOUR |
HD input (AVC >10mbps) | $0.2940 |
Output 1 (_416x234_150k) | $0.3540 |
Output 2 (_480x270_300k) | $0.3540 |
Output 3 (_640x360_600k) | $0.3540 |
Output 4 (_768x432_900k) | $0.3540 |
Output 5 (_960x540_1500k) | $0.3540 |
Output 6 (_1280x720_2475) | $0.7020 |
Output 7 (_1280x720_3750k) | $0.7020 |
Output 8 (_1280x720_4875k) | $0.7020 |
Output 9 (_1920x1080_6000k) | $0.7020 |
Total per hour | $4.87 |
This hourly cost is stable for this configuration.
Next, we’ll look at packaging and origination for the live channel using AWS Elemental MediaPackage. The first half of this cost is easy to work out, as we already have the information we need from the decision about encoding parameters.
OUTPUT | kbps |
Output 1 (_416x234_150k) | 150 |
Output 2 (_480x270_300k) | 300 |
Output 3 (_640x360_600k) | 600 |
Output 4 (_768x432_900k) | 900 |
Output 5 (_960x540_1500k) | 1,500 |
Output 6 (_1280x720_2475) | 2,475 |
Output 7 (_1280x720_3750k) | 3,750 |
Output 8 (_1280x720_4875k) | 4,875 |
Output 9 (_1920x1080_6000k) | 6,000 |
Total kbps | 20,550 |
GB/hour (20,550/8/1024/1024 x 60 x 60) | 8.82 GB/hour |
The total for data per hour for a one live channel output is 8.82 GB. MediaLive provides two sets of outputs per standard channel for redundancy, so the total output for the live channel doubles to 17.64 GB. The hourly cost for ingesting this live channel into MediaPackage in the US East (N. Virginia) region would be $0.7056 per hour (17.64GB * $0.04).
We now have the cost of encoding the live channel with MediaLive and ingesting it into MediaPackage as $5.576 per hour. If no one was watching the channel, we could stop here with our pricing. But that’s not practical, so now we have to move on to the areas that require estimations.
Adding in estimated costs for audience watching the stream
The cost of packaging, origination, and distribution are related to the volume of data being served, which is in turn related to the total audience for the live channel, and the quality (or bitrate) that they are viewing. To get to an answer that can be used to at least model and estimate cost, a target or prediction needs to be made. This can be informed by data from existing or past channels to see what the historic average audience size is, and the average bitrate that audience gets.
The prediction I’ll use is for one hour, with an average of 5,000 viewers watching at an average of 3 Mbps (25% saving over 4Mbps). This would mean a total egress of 6,591.80 GB per hour.
Viewers | 5,000 |
Average mbps per viewer | 3 |
Total MB per sec (5000 x 3 / 8) | 1,875 |
Total Egress per hour (1,875 /1024 x 60 x 60) | 6,591.80 GB/hour |
This means the cost for Amazon CloudFront distribution within the US would be $560.30 (6,591.80 * $0.085).
The cost for MediaPackage for packaging and origination in the US East (N. Virginia) region would be:
Using a 99% cache/hit ratio, $3.296 per hour. (1% of total egress 6,591.80 = 65.92GB * $0.05 = $3.296/hour)
What we have worked out so far – with iSIZE solution
We now have a more complete view of cost for a live channel with input of 15 Mbps 1080p AVC is encoded into the Apple-recommended TN2224 HLS parameters with AWS Elemental Media Live and ingested into AWS Elemental MediaPackage. MediaPackage formats and originates the video content, and Amazon CloudFront handles distribution, with a cache/hit ratio of 99%/x7.5 egress factor, to an average of 5,000 viewers watching at an average of 3 Mbps. The total cost in the US East (N. Virginia) region for this is $580.49 per hour.
AWS Service | Function | Cost per hour |
iSize Precoder | Precode Input/Output | $11.31 |
AWS Elemental MediaLive | Input and Outputs for Channel | $4.87 |
AWS Elemental MediaPackage | Ingest of channel | $0.71 |
AWS Elemental MediaPackage | Packaging and Origination | $3.30 |
Amazon CloudFront | Distribution | $560.30 |
Total | $580.49 |
What we have worked out so far – without iSIZE solution
AWS Service | Function | Cost per hour |
AWS Elemental MediaLive | Input and Outputs for Channel | $4.87 |
AWS Elemental MediaPackage | Ingest of channel | $0.94 |
AWS Elemental MediaPackage | Packaging and Origination | $4.39 |
Amazon CloudFront | Distribution | $747.07 |
Total | $757.28 |
This shows how easy implementation of iSIZE AI Precoder solution can save $176 per hour per 5,000 viewers of Live Streaming channel.
So, for a Live streaming customer with 10m viewers platform, assuming 1 hour per day average viewer time, iSIZE solution can save $352,000 every day.