LAMENT: A tale of constant struggle of what it's like trying to scale on ICP

Hi and thanks for the comment, I believe that this would be the ultimate goal, but to get there it might take very/too long (considering mainly existing projects which want to scale and run), to improve current projects/developers experience regarding scaling issues I’d suggest to consider (also) following as minimum:

  1. provide clear instructions for developers on how to do scaling on ICP
    • this should include (automated) procedure and instructions for moving canisters from overloaded subnets
  2. limit (number of canisters in subnets and) canisters usage of cpu/mem/iops/network/any-related resources - to prevent overloading of subnets (actual DDOS attack)
    • I don’t have insights of the current architecture, but this could involve also tuning of kernel parameters, cgroups, limits… on node machines - depends on what and how is actually running in the OS
1 Like