An SpmdEnabled pool cannot continue once communication between workers or between workers and client has been lost. If you are using the local scheduler or MJS and only using parfor then they are able to instead specify the flag 'SpmdEnabled' 'false'. See doc SpmdEnabled for details.
With this option the remaining workers will continue to complete the parfor even after 1 worker has lost connection.
If this is not what you were looking for, then it would help to understand if you could share the workflow, the cause of abort and the error messages you observe.