Skip to content
0:00 / 0:00

#lamportclocks#logicalclocks#vectorclocks

Paul Borrill on Time clocks and the reordering of events

PapersWeLove

23K subscribers • 34 videos on Hallway • Community channel

YouTube
  • Recommended by 1 person: 1 on Lobsters.
  • Presented by Paul Borrill

Intermediate. Assumes working knowledge.

4.9K views10 years ago1 recommendation

Distributed systemsHistory and cultureConsensus: Raft and PaxosReplication and consistencyHistory of computing

Paul Borrill reviews Leslie Lamport's seminal 1978 paper on logical clocks and the ordering of events in distributed systems.

Mini Adrian Cockcroft on Communicating Sequential Processes (spinroot.com/courses/summer/Papers/hoare_1978.pdf) Main Talk Paul Borrill on Lamport’s unfinished revolution. This talk reviews Lamport’s seminal 1978 paper on Time, Clocks and the Ordering of Events, the 2nd most cited paper in all of computer science. Almost all software engineers claim to have read it. Many who haven’t read it, use (and basically understand) the fundamental idea of logical clocks, and their progeny (vector clocks, matrix clocks, etc.). More than a few understand the current state of the art: dotted version vectors and bounded version vectors. Paradoxically, almost everyone missed some of the more subtle concepts, and questions that Lamport introduced in this paper. In the intervening years. Progress has occurred, and the state of the art has evolved. This talk is therefore in three parts. The first being a review of the paper itself, the concepts it introduced, and the assumptions behind these concepts. The second part reflects what we’ve learned in the intervening years, and especially the relationship of Lamport’s (original) understanding of time, which was superior to almost all other computer scientists at the time, and what (in contrast) we know now. The third part will be entirely devoted to questions and answers: Where anyone can ask a question, and anyone can try to answer it. The speaker will try to answer the question if no one else wants to, or if the audience appears dissatisfied. The discussion is expected to be lively, insightful, and potentially, mind blowing. In order to prepare yourself for this talk. Anticipate there will come a point where you are asked to take a blue pill vs. a red pill. If you watch these videos, your red pill transition will be gentler. Most of you may prefer to take the blue pill and go back to your old way of thinking about time. In which case, you won’t find this talk very interesting, because of course you already know all the answers. Those who want to be more prepared for the red/blue pill experience, will benefit from both reading the paper, and watching the most recent PBS videos on the nature of spacetime: 1. Does the Speed of Light have anything to do with light? youtube.com/watch?v=msVuCEs8Ydo PBS Digital Studios. A 12 minute and 45 second Red Pill: The Speed of Causality 2. Are Space and Time an Illusion/ youtube.com/watch?v=YycAzdtUIko An 8 minute and 54 second Red Pill. The Space Time Illusion 3. The original paper can be found here: research.microsoft.com/users/lamport/pubs/time-clocks.pdf Paul's Bio Paul Borrill Recently left the Infrastructure team at Apple. Previously, CEO of REPLICUS Research, REPLICUS Software, VP/CTO for VERITAS Software; VP/Chief Architect for Quantum Corporation; Distinguished Engineer, Director of Architecture & Performance, and Chief Scientist for IR at Sun Microsystems. Paul was Founding Chairman for the Storage Networ

Why it's here

1 person recommends this

tfon Lobsters • Recommends

Papers We Love SF: Paul Borrill on Time clocks and the reordering of events Paul Borrill on Lamport’s unfinished revolution. This talk reviews Lamport’s seminal 1978 paper on Time, Clocks and the Ordering of Events, the 2nd most cited paper in all of computer science. Almost all software engineers claim to have read it. Many who haven’t read it, use (and basically understand) the fundamental idea of logical clocks, and their progeny (vector clocks, matrix clocks, etc.). More than a few understand the current state of the art: dotted version vectors and bounded ver…