Dears, One hypothesis could be related to the configuration of the HTCondor scheduler and the way it allocates tasks to the available resources. It's possible that the scheduler might not be fully aware of the newly added machines or is not effectively distributing jobs across all available slots ? A misconfiguration or an oversight in the scheduler's setup might be causing this discrepancy ? Has anyone encountered a similar situation or can provide insights into what might be causing this behavior? Your expertise and suggestions would be greatly appreciated in resolving this matter and maximizing the efficiency of our HTCondor environment. |