...
The second round of tests were designed to load a larger number of containers. The initial Bash script was modified to use curl, instead of benchtool, to submit containers to Fedora 4. These tests were single threaded and attempted to load 100k, and then 500k, containers with 1 MB nonRdfSourceDescriptions binary into a flat fcrepo4 structure. These tests ran without exception, but showed a definite slow down in the containers' ingest times as the number of containers ingested continued to grow. The second test (of 500k containers) was discontinued at 175k containers when the increase in ingest times was determined to be growing to the point of being prohibitively slow. The graphs for these tests are near the bottom of the page (the last two red graphs).
...