Changes between Version 2 and Version 3 of Ticket #3952, comment 2


Ignore:
Timestamp:
2022-12-13T16:35:09Z (2 years ago)
Author:
itamarst
Comment:

Legend:

Unmodified
Added
Removed
Modified

  • Ticket #3952, comment 2

    v2 v3  
    11For a small 2000 byte file with a single share uploaded to a single node, an upload is 3 sequential HTTP requests, and a download is 4 sequential HTTP requests. So with high latency this gets quite slow.
    22
    3 Next, I tested over loopback, but with RTT latency of 60ms (30ms one way) artificially added. Bandwidth is quite high, so not likely to be a bottleneck.
     3Next, I tested HTTP over loopback, but with RTT latency of 60ms (30ms one way) artificially added. Bandwidth is quite high, so not likely to be a bottleneck.
    44
    55||= File size =||= Upload clock time =||= Upload CPU time =||= Download time =||= Download CPU time =||
    … …  
    1414* Immutable uploads use parallelism, and main bottleneck is latency. Upload time is not linear with size, . However, sequential uploads will have minimal time no matter what the bandwidth.
    1515* Immutable downloads have a massive latency hit, probably due to not utilizing parallelism well, so large files suffer from latency hit even when bandwidth is not a bottleneck. Also using a lot of CPU.
     16
     17I then tested Foolscap with 20MB file. Upload was similar, but download was 20,800ms clock time and 2,100 CPU time. So HTTP download is worse than Foolscap, but Foolscap downloads are still quite bad.