S3 vs. native usage comparison?

Hi all,

Has Storj published any information showing how much ingress and egress usage is being driven by S3 clients vs. native Storj API clients and which clients are being used (e.g. rclone, uplink, other client applications, etc.)?

I wouldn’t be surprised if most ingress traffic is via S3, but I’m curious if more people are using the native API for its performance benefits when downloading.

I didn’t find anything in the published stats or in the forum, so I figured I’d ask in case it had been mentioned and I missed it.

Thanks!

From the storagenode side it’s vividly obvious that virtually nobody is using native integration.

Top three sources and destinations for traffic for all of my nodes are (stats from the past few hours):

  • AS60068 (Datacamp Limited): 53.2% of traffic
  • AS212477 (RoyaleHosting BV): 27.0% of traffic
  • AS32475 (Internap Holding LLC) 9.5% of traffic.

Everyone else is in the remaining 10%.

2 Likes

Thanks for the response. That’s interesting.

Out of curiosity, what’d you use to parse out that data? I’m considering writing a script to parse my storage node’s logs and generate similar data, but it’d be nice to not have to reinvent the wheel.

“Hey codex, please write me a web service to monitor node performance, with the following features…”.. few evenings later I have this: https://nodes.arrogantrabbit.com

1 Like