The problem of medium to large file fuse concurrent reading

[jayzhenghan]

Problem1: Compared with single concurrent reading of the same file（300MB）, the delay of multiple concurrency increases dozens of times。fuse options(kernel_cache, max_readahead=32) 8 concurrency reads

5 concurrency reads

3 concurrency reads

1 concurrency read

Problem2 fuse options(kernel_cache) alluxio received out-of-order read requests from readahead of the kernel

Problem3 Serious read magnification。

1. 3 concurrent read 300MB.file（100% cache），
2. 3 concurrent read 300MB.file （no cache）

[jayzhenghan]

@LuQQiu @maobaolong @yuzhu

[maobaolong]

@jayzhenghan Would you please pasted the related resources here, something like read.py

[jayzhenghan]

@jayzhenghan Would you please pasted the related resources here, something like read.py

fd = open("test/300MB", "r") print fd line = fd.readlines() fd.close()

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed in two weeks if no further activity occurs. Thank you for your contributions.