You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Aug 2, 2021. It is now read-only.
This test would bring up a cluster of N nodes and would start filling up storage with equal size files f_0, ..., f_i. After that when uploading f_i+j, retrieval of f_j is attempted.
Report i+j, in which j is the aggregate network size, such that f_j is not available.
we have to do some kind of sliding window upload and report when older uploads are not available
we should specify “older” as a parameter - it is not clear what capacity the network has so to say
i have tried this manually multiple times but we have to automate it
i would go as far and say that we should not only attempt f_j on step f_i+j, but all f_0 to f_i+j-1
that would put them all in the caches or respective nodes though
Background: There is/was a bug where all chunks got sync to all nodes.
Depends on #1071
This test would bring up a cluster of N nodes and would start filling up storage with equal size files f_0, ..., f_i. After that when uploading f_i+j, retrieval of f_j is attempted.
Report i+j, in which
j
is the aggregate network size, such that f_j is not available.(i+j)*filesize*(MinProxBinSize+1) ~ node_count * storage_capacity
The text was updated successfully, but these errors were encountered: