There has been a lot of development going on in the Livepeer ecosystem over the past few months as the community works towards creating a useful, scalable, and cost effective open video infrastructure. Some of the areas of focus have included:
- Streamflow — the big network scaling update
- GPU mining and transcoding opportunity
- Video developer productization and pilot program
- Network participation and governance
- Token holder community product and growth
- This post shares some status updates, as well as provides a glimpse at what’s next in each of these areas.
Streamflow is the name of the upcoming
Livepeer protocol update focused on taking the initial alpha public network, and making it scalable, cost effective, and reliable. It is currently running successfully on an internal test network, and working with high reliable for video transcoding. Tests have shown up to 99.99% reliability with many video streams, under certain conditions, and the challenge will be to replicate this on a public test network before going to mainnet.
One of the interesting features of the Streamflow update is that it contains what we believe to be the first implementation of a layer 2 payment scaling solution called Probabilistic Micropayments, which enables users to pay providers on the
Livepeer network at high scale, without being effected by high fees on the Ethereum network.
Dashboard showing Probabilistic Micropayments tickets being sent, received, and redeemed for one node on the Streamflow internal testnet.
Another impactful feature of Streamflow is that it will enable increasing the number of active Orchestrator (currently called Transcoder) slots from 25 in the alpha, to 100+.
Audits for the Streamflow code are beginning the first week of October. The aim is for the public Streamflow testnet to launch in the final week of September or first week in October. The goal for the public testnet is to allow node operators to learn to play the roles of Orchestrator and Transcoder in the new protocol, to test out GPU transcoding, and to help find and report bugs to enable Livepeer to achieve 99.99% transcoding reliability before going to mainnet.
A long anticipated aspect of the Livepeer project was that it could enable existing crypto miners to continue to mine GPU based cryptocurrencies like Ethereum or ZCash, while simultaneously leveraging unused video encoding chips on their GPUs to make additional revenue transcoding video. GPU mining is now working in the Streamflow codebase, and the dual mining ability is also demonstrated, while benchmarks are being collected to report on tuning and performance impacts.
While income from transcoding will be unlocked with the Streamflow release, those looking to experiment with running both simultaneously can follow these early technical instructions. More polished software easing the miner experience will surely follow in the future.
Livepeer’s network is open, and it enables anyone to be able to tap into the protocol directly to get live video transcoded. However it has also become clear through working with the existing video industry, that more polished products and services would need to exist in order to let scaled existing video applications test and begin to integrate without worrying about blockchain, cryptocurrency, and decentralized tech.
The first product developed in this vein is a plugin to existing media servers that allows users to use the
Livepeer network for affordable and scalable transcoding, while hiding all of the crypto-complexity. Livepeer invites users of popular media servers to participate in a pilot program which will enable them to evaluate Livepeer’s technology at zero cost up to a limited amount of scale. The experience is that video developers can enable the plugin with a press of a button, and begin observing the benefits of Livepeer — affordable transcoding and reduced infrastructure costs.
These pilots are available to today, and we look forward to sharing positive case studies as part of the early usage. If you have a video streaming application in need of transcoding, please share your information here.
Participation in the public Livepeer alpha network has grown tremendously since launch. With 30+ infrastructure operators around the world running transcoder nodes, thousands of stakeholders participating in delegation to QA and secure the platform, and a large community of developers contributing to building tools around the supply side ecosystem.
The participation rate has recently passed the 50% target, meaning over 50% of all outstanding LPT is staked. The effect of surpassing this target is that the inflation rate now decreases by a small amount each round (day) that participation remains above this target.
Participation rate steadily climbing in 2019 to above 55%.
As the incentives to participate shift as a result of these economic parameters, it raises the questions around future governance. Livepeer’s road to decentralization summarizes the governance follows as the next stage in decentralization, and while forking to invoke an updated ruleset is always possible, there is a lot to be learned from recent protocols that have moved these sorts of upgrade mechanisms on chain and into community hands with seamless mechanisms involving signaling, proposals, voting, and binding on chain updates. This will be an active area of research and implementation for the project in early 2020.
As the most active set of participants in the Livepeer community are the token holders who have been operating the network, the project and community have put a lot of recent work into supporting this group.
