How does each Pod handle requests?

Question

Regarding Scalability field when creating a new HULFT Integrate Service, I understand from another FAQ that Pods are created as many as I choose for Scalability, and that availability can be ensured by increasing the number of Pods.

I have questions about request processing when there are multiple Pods.
*Assume that the each Pod are Pod-A, Pod-B, and so on.

1. What is the unit of measure for a request Integrate Service process?
   
   
2. When I run a script, are the requests always destributed to each Pod in like a load balancer?
   Or is there a function that allows Pod-A to handle all the requests and failover from Pod-A to Pod-B if something goes wrong with Pod-A, to Pod-C if something goes wrong with Pod-B, and so on?
   
   
3. If I create two Pods, Pod-A and Pod-B, in a Integrate Service and Pod-A stops for some reason, is it possible to continue processing subsequent requests only with Pod-B?

Answer

1. The request unit is a series of script processes from Start to End icons issued when a script is executed using specified Integrate Service.
   In other words, script execution is a single request.
   
   
2. When multiple Pods are created with Scalability settings, the requests are destributed by load balancer.
   Therefore, if the resource load of Integrate Service is high, it is effective to increase the number of Pods and resource size.
   
   
3. If Pod-A stops, the subsequent requests will be processed by Pod-B.
   If multiple Pods are created to increase scalability, even if one Pod stops, subsequent requests will be handled by the other running Pods.
   Therefore, availability can be ensured by increasing scalability.