rauljordan commented 6 years ago

Prysmatic Labs Proposal: Sharding Client Implementation for Go-Ethereum

Abstract

Currently, a variety of off-chain scaling solutions for Ethereum are in the works, from Plasma to Generalized State Channel implementations. These approaches advance the scalability of the protocol, but do not provide a solution to the increasing computational and storage requirements for nodes. More specifically, off-chain scaling does not solve the problem of an ever-increasing main chain that fewer and fewer consumer devices are able to store. The mounting cost for full nodes will eventually become economically infeasible; especially for mainstream adoption.

Sharding is an layer-1, in-protocol modification to Ethereum that provides a reasonable solution to the scalability trilemma.

In this scheme, we partition the blockchains into k shards, where each shard has O(c) processing power, allowing the network to process transactions in parallel. Sharding for Ethereum is a scalability approach that does not introduce additional computational requirements on nodes. The difficult part of implementing this system is preventing degradation of security in the network, as new attacks such as 1% shard takeovers now exist.

At Prysmatic Labs, we have assembled a strong team of open source developers working on implementing a sharding client for Ethereum through a strong focus on being practical, transparent with the developer community, and constant interaction with Ethereum researchers. We have made strong progress so far, and have put together clear documentation outlining our work.

Deliverables

Sharding is a long, difficult research effort. However, Prysmatic Labs has 3 milestones in mind within its scope:

Our first release will be a working demonstration of the entire, end-to-end sharding system for anyone to try on their personal computer as a Geth client. We are currently making progress towards this milestone as tracked by our project board.
A working version of a geth sharding client that interacts with a Sharding Manager Contract deployed on the Ethereum public testnet or another sharding specific, public testnet.
A working version of the geth sharding client on the Ethereum mainnet. This will be a large, multi-team effort that will require substantial research and time invested in it.

Grant size

Funding: $100k in ETH, paid as separate chunks, one before our first release and the rest after our first release, as the testnet deliverable will be a much longer, time intensive effort.

Success reward: Up to $50k in ANT, given out when public testnet release is ready

Application requirements

-Proof of concept of our current work so far and how to run it on a local computer. -Details of the core team members -Details of all our public communication channels (Website, Gitter, Medium page updates, Twitter, Email Newsletter, Youtube recorded team meetings) -Details of where we currently stand on the research for our next steps

Development timeline

The development timeline will be the following one in regards to each deliverable:

Late Summer - Early Fall 2018
Q2 2019
Q4 2019 - Q1 2020

mariapao commented 6 years ago

Hi @rauljordan thanks for submitting your proposal. We like the project and the team a lot and have decided to consider your proposal for a grant. The next step is to submit a request for funding which I see you already did :) I, together with @bingen, will review your request for funding and give you feedback during the next couple of days.

prestonvanloon commented 4 years ago

To give an update here. Our team was notified today by the Nest Program Manager, @burrrata, that we will not be receiving the success reward of "Up to $50k in ANT". The reason given is that the Nest program has cut back funding due to the bear market.

Our testnet was launched in Q2 of 2019 and has gone through many iterations of improvements. At the moment, it is the largest ETH2 testnet with 35k+ validating keys and over 100 nodes in the network. By most standards, we consider this testnet a wild success and we are disappointed that the Aragon Nest program would not honor any ANT token success reward.

prestonvanloon commented 4 years ago

The conversation was in a private direct message. I don’t mind sharing the conversation if @burrrata is OK with that.

jpitts commented 4 years ago

@prestonvanloon, If you are going to proceed with an open approach to resolving this matter e.g. sharing commitments made, chat messages, etc. could you consider doing it in a new GitHub repo? This repo could contain your complaint, evidence you submit, their response, evidence they submit, and so on. Using a repo puts the relevant info all in one place, and with data integrity.

It is unknown what would happen later in terms of your situation being resolved, but the approach for arbitration could be agreed upon later.