Do a simple FOAF crawl through LinkedDataSail and observe the thread ids in
Ripple's log. Thirty threads are available by default, but for long
periods, only one to three threads will be active at a given time. What's
the problem with the other threads? Are they hanging, is LDS requesting
new threads less than it should, or is the scheduler handing out new
threads less than it should?
Original link: http://code.google.com/p/ripple/issues/detail?id=45