Is there any way of extracting questions from a big PDF without overloading the server ?

I'm trying to  extract  a  big amount of question from a  PDF file that has around 1000 pages  and the section  where we find all those questions.

The program is doing it  we can run  the process an everiything is fine but suddenly  the appian server  shows the 503  error   even though we are not running the process i split the pdf  in  batches of 30 pages  every time a 30 page batch   is evalueated and the  questions are extracted we save the   questions found in this  variables and this sub process die and  go  to the main process  and do the same  thing till  we get to  have all the questions and sections   the  time for this process is about 10 minutes in doing all of this operations.

how can we minimize the high load of the site ?

 the Appian team told us that the aplication server was killed by their automation. After failing multiple readiness check  and

that the metrics being exported from the application server stopped being reported which indicates unhealthiness.

This  503 error  happend without any process being runing