Nova-conductor processes access the database directly. Since the database connection driver is typically implemented in a library beyond the purview of eventlet’s monkeypatching (i.e., a native python extension like _mysql.so), blocking database calls will block all eventlet coroutines. Since much of what nova-conductor does is access the database, a nova-conductor process’s handling of requests is effectively serial.
To mitigate this problem, you can simply run more nova-conductor processes. Deploying multiple conductor workers per host avoids serialization on database accesses caused by libmysqlclient.so blocking eventlet's single thread. In an experiment on a 24-core machine, when creating 20 VMs in parallel, maximum creation time was reduced by approx. 10s when using 20 conductor processes vis-a-vis a single conductor process. Profiling showed that all of the savings came from faster calls into nova.db.sqlalchemy.api.
To make running multiple nova-conductor processes straightforward, there should be a workers=N option in the [conductor] section of nova.conf -- just like the osapi_compute_workers=N flag in the [DEFAULT] section.
Makes sense. Are you planning to submit a patch?