Source
The meeting itself — found from sitemap — Jitsi Meet
Checked for Jitsi Meet on 1 Oct 2026
- Page
- https://jitsi.org/jitsi-videobridge-performance-evaluation/
- Checked
- 1 Oct 2026, 12:12 UTC
- How we may use it
- Public page, crawling permitted
Technical details
- type
- page
- http status
- 200
- content hash
- sha256:511eb7aa96900627b10ffb1248b287bc810e95203ec91217161fd8f9e3adc9ab
- permission
- robots_ok
- screenshot
- Screenshot on file (internal exhibit, not published)
Cited by
Facts read from this source
-
Performance summary Report an error
“On a plain Xeon server (like this one) that you can rent for about a hundred dollars, for about 20% CPU you will be able to run 1000+ video streams using an average of 550 Mbps!”
-
Test server cpu Report an error
“The server machine is equipped with a quad-core Intel ® Xeon® E5-1620 v2 @ 3.70GHz CPU.”
-
Test stack Report an error
“The server is running all the services needed by Jitsi Meet: a web-server (nginx), an XMPP server (prosody) and Jitsi Videobridge.”
-
Test stream count Report an error
“In this run we used thirty-three load generators and caused Jitsi Videobridge to distribute 1056 streams (i.e. 1056 video and 1056 audio streams). Bandwidth usage reached over 550Mbps.”
-
Cpu usage at max streams Report an error
“We see how the 1056 streams end up taking about 20% CPU.”
-
Equivalent conferences Report an error
“This generates a bitrate roughly equivalent to 528 one-to-one conferences, 176 3-person conferences or 53 conferences with 5 people.”
-
Process rss Report an error
“during our tests, the RSS of the Jitsi Videobridge process as reported by ps(1) never exceeded 1500MB.”
-
Load generator bitrate Report an error
“each streaming a pre-recorded audio and video file with a combined average bitrate of about 515 Kbps per endpoint.”
-
Scaling Report an error
“The number of actual video streams increases quadratically with the number of load-generating endpoints”
-
Feature last n Report an error
“Usually, for large-scale conferences, people would want to use Jitsi Videobridge’s Last N mode so that only the most recent speakers would be distributed and shown.”
-
Cpu linear with bitrate Report an error
“the CPU usage increases nearly linearly with the total bitrate.”
-
Benchmark 90 streams Report an error
“90 Streams Bitrate mean: 47.6Mbps CPU usage mean: 3.1%”
-
Benchmark 380 streams Report an error
“380 Streams Bitrate mean: 199.4Mbps CPU usage mean: 8.0%”
-
Benchmark 600 streams Report an error
“600 Streams Bitrate mean: 314.7Mbps CPU usage mean: 11.7%”
-
Benchmark 812 streams Report an error
“812 Streams Bitrate mean: 425.5Mbps CPU usage mean: 15.7%”
-
Benchmark 1056 streams Report an error
“1056 Streams Bitrate mean: 550.4Mbps CPU usage mean: 20.3%”
-
Authors Report an error
“by Boris Grozev and Emil Ivov”
-
Trademark holder Report an error
“Jitsi is a trademark of © 8x8, Inc. All Rights Reserved.”
1 fact read from this page is not shown because it could not be confirmed on the page as captured.
Scores citing this record
- The Accessibility Advocate The meeting itself
- The Accessibility Advocate Pricing transparency
- The Educator The meeting itself
- The IT Administrator The meeting itself
- The IT Administrator Pricing transparency
- The Security Officer The meeting itself
- The Security Officer Reach & accessibility
- The Skeptic The meeting itself
- The Works Council Advocate The meeting itself