How Google designed their Highly Available Load Balancers using Anycast

Поделиться
HTML-код
  • Опубликовано: 20 окт 2023
  • System Design for SDE-2 and above: arpitbhayani.me/masterclass
    System Design for Beginners: arpitbhayani.me/sys-design
    Redis Internals: arpitbhayani.me/redis
    Build Your Own Redis / DNS / BitTorrent / SQLite - with CodeCrafters.
    Sign up and get 40% off - app.codecrafters.io/join?via=...
    In the video, I discussed a research paper from Google titled "Anycast as a Load Balancing Feature" published in November 2010. The paper explores using anycast for load balancing and failover services, providing insights into design decisions made at the time. It delves into the architecture of Google's subsystems and how anycast was implemented to offer failover services. Various solutions such as DNS-based load balancing and anycast routing technique were explored to ensure seamless service redirection in case of failures.
    Recommended videos and playlists
    If you liked this video, you will find the following videos and playlists helpful
    System Design: • PostgreSQL connection ...
    Designing Microservices: • Advantages of adopting...
    Database Engineering: • How nested loop, hash,...
    Concurrency In-depth: • How to write efficient...
    Research paper dissections: • The Google File System...
    Outage Dissections: • Dissecting GitHub Outa...
    Hash Table Internals: • Internal Structure of ...
    Bittorrent Internals: • Introduction to BitTor...
    Things you will find amusing
    Knowledge Base: arpitbhayani.me/knowledge-base
    Bookshelf: arpitbhayani.me/bookshelf
    Papershelf: arpitbhayani.me/papershelf
    Other socials
    I keep writing and sharing my practical experience and learnings every day, so if you resonate then follow along. I keep it no fluff.
    LinkedIn: / arpitbhayani
    Twitter: / arpit_bhayani
    Weekly Newsletter: arpit.substack.com
    Thank you for watching and supporting! it means a ton.
    I am on a mission to bring out the best engineering stories from around the world and make you all fall in
    love with engineering. If you resonate with this then follow along, I always keep it no-fluff.
  • НаукаНаука

Комментарии • 18

  • @anirbanhaldar7081
    @anirbanhaldar7081 8 месяцев назад +9

    Wow.. after watching Dukaan podcast .. I was having some doubt how their failover is happening.. good to see thats covered here in detail ❤

  • @koteshwarraomaripudi1080
    @koteshwarraomaripudi1080 8 месяцев назад +3

    Great initiative to go through a white paper and explaining us. Keep the content coming. Thanks a lot. Hoping to see more of these discussions on Distributed systems/DB papers

  • @santoshreddy9628
    @santoshreddy9628 8 месяцев назад +1

    Thank you Arpit its very interesting concept and paper. Appreciate the time you are giving and knowledge you are sharing. I recommend your vidoes to my team they enjoy learning from you. Thank you.

  • @PoojaDurgi
    @PoojaDurgi 5 месяцев назад

    Your'e the Best teacher Arpit !!! Thanks so much.

  • @daegu_1
    @daegu_1 8 месяцев назад

    Thanks for great explanation

  • @shubhamgupta-nc1vx
    @shubhamgupta-nc1vx 6 месяцев назад +2

    9.5k views only? This content is pure gold man!

    • @AsliEngineering
      @AsliEngineering  6 месяцев назад +3

      Mai bhi yahi sochta hu 😅 nahi dekhna logo ko deep tech. Motivation and roadmap dedo bas.

    • @yashkhd1100
      @yashkhd1100 6 месяцев назад +1

      @@AsliEngineering That is unfortunately the case. In general a very common topics(even c/c++ despite having tons of videos) get lot of views.

  • @nishantnasa8798
    @nishantnasa8798 6 месяцев назад

    Thanks, I was interested in this as well after watching Dukaan podcast. Can you share what tools you use for video creation? I am interested in sharing some tech content myself & have been inspired from your journey. Also one appreciation that you zoom in on diagrams during explanation, that's really helpful for me as I have low vision so keep doing that too.

  • @ankuragarwal9712
    @ankuragarwal9712 8 месяцев назад

    What does advertising actually mean? Can you please provide specific term so that I can search about it and read?

  • @sanghamitrahota1162
    @sanghamitrahota1162 8 месяцев назад

    Hi, before going through a paper, can you post in advance, so that we would also skim it through in advance . If you are already doing this, I am not aware of it. Please let me know where I can find the posts in ?

  • @guytonedhai
    @guytonedhai 8 месяцев назад

    So even if 1 service under the load balancer goes down, we make the load balancer undiscoverable? Is my understanding correct?

    • @AsliEngineering
      @AsliEngineering  8 месяцев назад

      No. All the services. Google made the change to LDirectorD and added fallback command which does this.
      When none of the services are working it stops advertising itself as anycast peer.

  • @sonusuresh6642
    @sonusuresh6642 8 месяцев назад

    I like everything about your channel. except the hardcoded captions/subtitles on your video 😝

  • @reena2485
    @reena2485 8 месяцев назад +1

    When LB stops advertising IP then who updates the routing table? That path from router to LB need to be removed. When failed site is up then LB advertise the IP, path need to be added back in routing table, how does that happen and who does it ?

    • @AsliEngineering
      @AsliEngineering  8 месяцев назад

      LDirectorD

    • @rahulsangvikar7973
      @rahulsangvikar7973 Месяц назад

      @@AsliEngineering Since anycast doesn't perform healthchecks, how does it know that the LB has stopped advertising the ip so that it can update its routing table?