Save on CDN expanses or reduce service load by utilizing MESH delivery
Ensuring fast and stable streaming is a must for any video oriented service. Some services setup their load balancing clusters and software while others offload these work by opting to use a Content Distribution Network service. Whatever the path you have chosen, Peerplug MESH Streaming can help you save valuable money and resources.
CDN pricing usually calculated by the amount of traffic delivered. Therefore, as your service grows in popularity CDN costs can add up to a hefty sum each month.
As CDN price is “per GB delivered”, CDN costs can be a major factor in deciding on the quality of your content. One of the core factors of video quality is it’s bitrate. Bitrate is how much “zeros or ones” per second the video requires. Here is the table describing the bitrate recommendation for popular resolutions:
From the table above, you can see that it is almost 4 times more expansive to deliver 4K video compared to delivering 720p video. Therefore if you decide to upgrade your service from 720p to 4K you may incur CDN cost of up to 4 times higher. Solution ? MESH streaming.
Peerplug MESH Streaming engine offloads the video from the CDN to the users. As users seamlessly share your content between themselves your CDN is only used during “cold start” and when there is a very low amount of users. Peerplug achieves 95% peer-to-peer offload rate in production environment. Meaning your CDN bill will be only 5% of you pre-peerplug bill.
You can use those resources to improve and future-proof your service by opting for higher bitrate increasing service quality or turn the saved funds into pure profit.
As an oversimplified example of CDN saving calculation, let’s assume your current CDN traffic price is 1$ per GB and you use 100GB per month. This results in a CDN bill of 100$.
We will assume Peerplug offloads 80% of your traffic usage to peer-to-peer and the cost of 1 GB you receive from Peerplug is 0.3$ (30% of your CDN). For CDN based customers, Peerplug always commits to traffic prices of no more than 30% of your CDN price.
After adding Peerplug to your service, your bill will be:
20% of your traffic (20GB * 1$) is still delivered by your CDN - 20$
80% of your traffic (80GB * 0.3$) is delivered using Peerplug - 24$
Total bill: 20$ + 24$ = 44$ instead of 100$.
56% more cost effective in addition to more stability and all other benefits of MESH streaming model.
Peerplug daily result in production environment of one of our early adopters:
For exact Peerplug CDN based pricing and free trial, please contact us.
Self hosted OTT services:
This approach is more cost effective for large volumes of traffic than using CDNs. This is especially true for IP Television service providers as amount of data transferred per month can be extremely high.
Self hosted OTT services reach the same limitations on video bitrate but instead of being limited by the CDN price, self hosted service is limited by the total bandwidth available.
Here is a two-week bandwidth usage history of a self hosted OTT service that has a prime-time at around 21:00:
Service should be able to provide at least 400 gbps of bandwidth. To ensure quality and be ready for a sudden spike as a result of a popular event, it is recommended to able to coupe with at least 600 gbps (+50%).
Thanks to Peerplug, this customer’s service infrastructure should only be able to coupe with about 60 gbps to provide the service and the rest is provided by Peerplug MESH streaming. In the graph above, the red part is what comes from service backend, blue part is Peerplug offloaded peer-to-peer traffic.
When using Peerplug, sudden spike in service popularity is not a problem at all, as it is well known that as there are “more peers” peer-to-peer networks work more efficiently.
How much do you save by using Peerplug can wary depending on how much does it cost for you to host
60 gbps up-link compared to 600 gbps. If your up-link pricing is linear, you save 90% of your infrastructure expanses!
Peerplug pricing for self hosted OTT services goes by either the amount of active SDK instances in total or by the maximal amount of simultaneously streaming users. The suitable pricing model will be selected depending on the type of the service and it's userbase.
For details and free trial, please contact us.