Systems Management

Ever since we updated to 7.0 at the end of last year, machines are taking a ridiculous amount of time to run detect jobs in our environment.  I have bumped the detect timeout up to the max of 5 hours and machines still time out or come back with errors after a job has run.  It is becoming impossible to run detect/deploy jobs as the machines never get to the deploy stage.  Support tells me we're trying to do too many machines or too many patches but its the same number we've always done and even when running a manual job on just a handful of machines (or even just one) we get the same results.  The 7.1 update seems to have improved things somewhat but we're still running into more of the same issues.  Any idea how to troubleshoot this? I've had an ongoing support ticket for a while and I'm not 100% sure how to proceed.