TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

A deep dive on load balancing algorithms

78 pointsby khalidlafiover 3 years ago

2 comments

jedbergover 3 years ago
I'm disappointed that this doesn't include HRW (Highest Random Weight or Rendezvous Caching). It has the advantages of consistent hashing without the disadvantage, and doesn't require a central coordinator like consistent hashing does.
评论 #29260436 未加载
recursiveover 3 years ago
I&#x27;m not familiar with any of these, but this point stuck out to me.<p>Two random choice says &quot;Max load of any server is O(log log number_of_servers).&quot;<p>If work accomplished is proportional to load, then the total work done by the entire system is O(number_of_servers * log log number_of_servers). It seems very suspicious, magical even, that the total work is more than linear with the number of servers. Free energy discovered?
评论 #29260221 未加载
评论 #29260500 未加载
评论 #29264639 未加载
评论 #29260199 未加载