The Stellar Consensus Protocol | David Mazières | Talks at Google
Talks at Google2 HN recommenders · recommended in 2 months, 2018-2021
56K views · 10 years ago · talk · evergreen
David Mazieres presents the Stellar Consensus Protocol at Google, introducing federated Byzantine agreement. The model lets participants choose whom to trust rather than assuming fixed membership.
Recommended as the protocol designer's own detailed explanation of SCP.
34:52Liberating Distributed Consensus - Heidi Howard
Reactive Summit
7K views • 7 years ago
1:06:34Paxos lecture (Raft user study)
Diego Ongaro
114K views • 13 years ago
1:19:35Lecture 1: Introduction
MIT 6.824: Distributed Systems
921K views • 6 years ago
58:17Raft lecture (Raft user study)
Diego Ongaro
87K views • 13 years ago
54:44Differential Synchronization
Google TechTalks
52K views • 17 years ago
43:39CRDTs and the Quest for Distributed Consistency
InfoQ
68K views • 8 years ago
48:02BlackHat USA 2011: SSL And The Future Of Authenticity
Christiaan008
95K views • 15 years ago
1:10:23"Systems that run forever self-heal and scale" by Joe Armstrong (2013)
Strange Loop Conference
85K views • 5 years ago
14:36Distributed Systems 1.1: Introduction
Martin Kleppmann
382K views • 6 years ago
46:15AWS re:Invent 2014 | (SPOT302) Under the Covers of AWS: Its Core Distributed Systems
Amazon Web Services
10K views • 11 years ago
56:19MINIX 3: a Modular, Self-Healing POSIX-compatible Operating System
FOSDEM
90K views • 16 years ago
46:22DEF CON 19 - Moxie Marlinspike - SSL And The Future Of Authenticity
DEFCONConference
11K views • 13 years ago
51:36Conflict Resolution for Eventual Consistency • Martin Kleppmann • GOTO 2016
GOTO Conferences
46K views • 9 years ago
43:0336C3 - The ecosystem is moving
media.ccc.de
2K views • 6 years ago
1:06:45Let's #TalkConcurrency Panel Discussion with Sir Tony Hoare, Joe Armstrong, and Carl Hewitt
Erlang Solutions
13K views • 7 years ago
1:10:10CRDTs: The Hard Parts
Martin Kleppmann
90K views • 6 years ago
48:45Matrix 2.0 — How we're making Matrix go voom!
Matrixdotorg
16K views • 3 years ago
48:29DEF CON 30 - Sam Bent - Tor - Darknet Opsec By a Veteran Darknet Vendor
DEFCONConference
557K views • 4 years ago
42:20"Stop Rate Limiting! Capacity Management Done Right" by Jon Moore
Strange Loop Conference
41K views • 9 years ago
51:27Google I/O 2012 - Go Concurrency Patterns
Google for Developers
894K views • 14 years ago
3 comments from Hacker News
It seems so. Its the notion of "Trust Lines"[0] All this thing does is limit potential losses from fraud. Not eliminating. The main value proposition of a blockchain is to solve the "principal-agent" problem [1] and gis simply reduces the risk but doesn't remove it entirely. Plus some of the requirements on a node having to be online (in a decentalized, byzantine environment) all the time are unrealistic. This seems more like ripple. Stellar uses a more sophisticated notion of "quorum slices" and is resistant to byzantine faults [2] [0]: link [1]: link [2]: ▶ this video
David Mazières (designer of the protocol) has given a number of talks. Here's a good video of his talk at Google, and I believe it's the one I watched a couple years ago: ▶ this video
Crap. I'll reproduce the answer here, then: > Tl;dr: Stellar’s protocol (SCP) uses a decentralized state propagation where each node commits to a value if every node it deems trustworthy agrees while Ripple uses a supermajority vote (80%) among all validating nodes. Stellar achieves better decentralization than Ripple > A pretty old, but still relevant discussion on the bitcoin stack exchange [ link ] > Stellar started out as a fork of Ripple (now completely rewritten) so they are pretty similar in implementation and use cases. > Stellar uses the Stellar Consensus Protocol [ link ] (SCP) to achieve decentralization, while Ripple uses the XRP Ledger Consensus Process [ link ]. These are alternatives to Bitcoin's proof of work whose goal is to have every participant in the network agree on the current state. > In SCP, each node decides for itself which other nodes it trusts. When a lot of these sets of trusted nodes are overlapping, there is a high probability there is agreement on the state of the network. Broadly, a node will only accept a message if all of their trustworthy nodes have in turn accepted it. This is similar to the way internet certificates and DNS servers are working. In principle, there is no central authority appointing trustworthy nodes, but de facto, the network will reach an agreement on which nodes are trustworthy. This is a great video explaining SCP in detail [ ▶ this video ]. There is also the white paper [ link ] for more formal mathematical defini