Today, we have open-sourced our implementation of a fully functional Loopring backend: something we call Lightcone relayer version 1.0. Here is a link to the GitHub repository: http://github.com/Loopring/lightcone.
A relayer is what hosts, handles, and matches orders, ultimately passing them through to the Loopring protocol smart contracts for settlement. Combined with a Wallet (frontend), it is what makes up a DEX. A single person/project can operate both a Wallet + Relayer (a DEX), or choose to build and focus on one of the two, working with a partner to provide the other function.
The Loopring Foundation built Lightcone 1.0 to:
- Show the ecosystem our interpretation of a Relayer to help developers get started building their own (it is not easy to build!)
- Allow projects (namely, Wallets) to hook up to Lightcone directly to access ‘Matching-as-a-Service’, so they can focus on their frontends
Ultimately, #1 has come true as we open source the code right now after ~8 months of building, but #2 will not proceed as planned. The Loopring Foundation has decided against operating Lightcone 1.0 as a matching engine service, and will not continue to develop this version.
Please read on for the reasoning behind this, and what’s in store for the future.
About Lightcone Relayer 1.0
Lightcone Relayer 1.0 is a backend matching system that supports Loopring Protocol 2.0. Lightcone 1.0 is fully functional, and test coverage is very high: we have been performing integration testing on Ethereum testnet for several weeks.
We believe that the source code of the Lightcone relayer can provide a very valuable reference implementation for community developers.
Importantly, as an off-chain system, much of the Lightcone relayer code is also suitable for other blockchains besides Ethereum. PoS-based public blockchains or Ethereum high-performance sidechains would do very well with a backend matching system like Lightcone because gas fees are typically lower for those chain types.
Why not operate Lightcone Relayer 1.0?
The matching efficiency of Loopring Protocol 2.0 is limited by the throughput constraints of Ethereum.
At the beginning of the year, our goal was to launch the Lightcone relayer and operate it with a base level of liquidity to help partners building on Loopring 2.0 to bootstrap liquidity. However, the progress of Loopring protocol 3.0 has exceeded our greatest expectations, and we’ve decided to not devote further resources on version 2.0, but focus on designing and developing protocol 3.0.
Currently, the gas required for settling a trade (order-ring) on protocol 2.0 is $0.15, and the throughput is about 2 trades per second.
The current 3.0 version can settle >450 trades per second (without on-chain data availability), and 80 trades per second (with on-chain data availability), and cost between $0.0004 to $0.003 per trade.
With furthers optimizations of the protocol, we believe that Loopring Protocol 3.0 will easily supplant 2.0-based DEX products and user experiences. With 3.0, we will bring fully-functional decentralized exchanges a step closer to competing with their centralized counterparts.
About Lightcone Relayer 2.0
With this in mind, we have started designing and developing Lightcone Relayer 2.0 — the version that will be compatible with Protocol 3.0 (we know, these numbers can get confusing!).
Basically, Lightcone 2.0 will seek to do the same as the first generation was meant to: 1) a full reference for developers to learn/build from, and 2) our previous concept of providing code or cloud services (Matching-as-a-Service) to select strategic partners.
We hope to have Lightcone 2.0 up and running by June 2019, about the same time that Protocol 3.0 will be deployed to testnet. Then, when Protocol 3.0 is deployed to Ethereum mainnet shortly after, we will have the full suite solution, and we believe, the world’s most powerful protocol and infrastructure for building non-custodial exchanges. Safe to say, we can hardly wait!
Back to the original point of this post: a powerful backend matching system is now open for anyone to build on, use, fork, modify, or learn from!
While Lightcone 2.0 coupled with Loopring protocol 3.0 is our focus, Lightcone 1.0 is formidable, and can be used as a backend system for Ethereum DEXs built with Loopring protocol 2.0. It is also suitable for the other chain types alluded to above, given less constraining throughput and costs. Keep in mind, Loopring 2.0 supports security tokens, so it can be used to build security token exchanges just as easily.
Do not take our forward-looking stance as a statement to its obsolescence — we believe it is surely among the highest performing order matching systems around today. We invite anyone to get building with it, or tweaking it for their own use.
To stay up-to-date with Loopring, please sign up for Loopring’s Bi-Weekly Update, and find us here:
⭑ Twitter: twitter.com/loopringorg
⭑ Reddit: reddit.com/r/loopringorg
⭑ Telegram: t.me/loopring_en & t.me/loopringfans (Chinese)
⭑ Discord: discord.gg/KkYccYp
⭑ GitHub: https://github.com/Loopring
⭑ Kakao: open.kakao.com/o/gJbSZdF (Korean)