Proof-of-Replication, Power-Fault-Tolerance and Research Roadmap
At Protocol Labs we are charting a course to the future, but we have!). At the same time, we’re also more than willing to openly discuss the challenges of implementing such a system. In fact, since the original Paper was posted in 2014, the team has been working on building various logistical and cryptographic components necessary for decentralizing the internet’s storage. In this post, we introduce two of those components: Proof of Replication and Power Fault Tolerance.know that the details make or break the plan. We love thoroughly explaining the reasons we think a decentralized storage network is worth building (and
IPFS is the largest and most visible component that we have constructed at Protocol Labs so far. This open source is built on IPLD and libp2p (also by Protocol Labs) and serves to efficiently address, locate, and transfer content from peer to peer, as an alternative to the classical (HTTP) web. This protocol is a requisite for a decentralized storage network, and continues to be improved and adopted by an increasingly diverse collection of decentralized applications and data archival groups.
Once we have built an efficient addressing and distribution protocol, it’s clearly necessary to incentivize people to host these files. Our aim has been to develop the most secure and robust reward system while still keeping file transfers efficient. We’ve summarized the results of this effort in the new Filecoin Paper, described here. Along the way we had to solve multiple self-contained subproblems within the protocol. Some of this work was done in collaboration with various academic experts at the forefront of cryptography.
In the spirit of open source, we’re posting this blog post with a shortand two technical reports on some of those subproblems. The following documents illustrate our progress to date in developing solutions. We will expand and improve upon them over time.
This document opens with a description of how far Filecoin has come and then describes the results that have been presented this year. We consider the core of the Filecoin protocol design to be stable, though we are open to improvements on some of its pieces.
The remainder of the Research Roadmap is a list of known future work and remaining open problems to be tackled in the remainder of 2017: finishing these papers, implementing Filecoin, making progress on outlined future work, and searching for solutions to a set of open problems.
Research Roadmap 2017
Problem: Clients will require storage providers to store multiple copies of the same data for reliability and availability. Attackers may attempt to get paid for storing multiple copies of data when in reality they only have a single copy. We need miners to prove they’re storing each and every copy of the data they claim they’re storing.
We developed the Proof-of-Replication, which solves this problem and a number of other malicious miner attacks. This involves a new kind of Proof-of-Storage where each physical copy is unique.
Problem: In the standard conception of Byzantine Fault Tolerance, various parties must reach consensus in a protocol where influence over the procedure is distributed equally and discretely. This does not accurately model consensus protocols based on Proof-of-Work, Proof-of-Stake, and others.
We present a formalism of Power Fault Tolerance, which cleanly generalizes Byzantine Fault Tolerance to situations where influence is distributed continuously and unequally.
Power Fault Tolerance
We welcome your feedback, collaboration, and conversation: firstname.lastname@example.org
点对点科技深耘IPFS与Filecoin技术，坚持区块链技术改变未来的信念。点对点 IPFS 数据中心是目前国内技术领先，性价比高、保障优的投资标的。自建杭州数据中心，合作数据中心分布于上海、宁波、河北、香港、斯德哥尔摩（瑞典）等地。点对点数据中心具有优秀的硬件配置与目前国内优质的网络节点资源。点对点科技力求将IPFS爱好者升级为IPFS领军者与受益者，让IPFS颠覆传统互联网，共同开启 WEB 3.0时代。