Join GitHub today
GitHub is home to over 36 million developers working together to host and review code, manage projects, and build software together.Sign up
Optimize leastActiveSelect and weight test case #2172
@@ Coverage Diff @@ ## master #2172 +/- ## ============================================ - Coverage 55.05% 55.01% -0.04% - Complexity 5282 5288 +6 ============================================ Files 573 573 Lines 25490 25538 +48 Branches 4526 4531 +5 ============================================ + Hits 14033 14050 +17 - Misses 9362 9387 +25 - Partials 2095 2101 +6
This was referenced
Aug 3, 2018
I saw your pr, I think the start warm up feature with the leastactive algorithm for loadbalance will be conflicted, because of the active for the new provider will be the least.
So is it necessary to calculate the weight for warm up?
Or maybe we can consider another implementation to supports warmup for leastactive loadbalance.
This is not the problem that this PR has to solve. The problem that this PR has to solve is too many loops and a small bug.