Inconsistency between time taken for write to db node to complete and the timing shown in the 'data-store-details.csv' log

chetany

A Score Level 1

over 8 years ago

Hello Everyone,

I have a query regarding the time of Write to Datastore operation reported in the log file "data_store_details.csv"
This log file breaks the total time for the write to DB query to execute in "Mean Prepare Time", "Mean Execute Time" and "Mean Transform time". Now, for one example - on one environment I am working on - the durations reported for these are 5ms, 16ms and 5ms respectively - for a total of 26ms of Mean Total time.

Now, I observed the time taken for the "Write to Datastore" node to complete in the process history of the process instance. It shows the time as 1.04 seconds. This indicates the Write to datastore node took 1.04 seconds to complete. However, from the log file - I see that the actual database operation took a total of 26ms.

So, why is there some...

Discussion posts and replies are publicly visible

Parents

0 aloks0189
Certified Lead Developer
over 8 years ago

@chetany i think process history holds overall time taken for this instance to get complete, rather than a particular node execution time, please correct me if i am wrong.

Also can you please let us know, how exactly are trying to capture the start and time and the difference between start and end time after the completion of Write to Data Store Node execution.

Yes, i agree with you that, Appian has to follow lot many set of steps before performing CRUD operation on DB. As Appian internally uses ORM (JPA, Hibernate) and ORM internally uses JDBC, so as per the JDBC internal mechanism, it has to follow various steps in order to perform any Operation on Relational Tabular Data store, these are:

1. Obtain connection to DB
2. Get a JDBC Statement
3. Execute the Statement
4. Close the Connection

Now when we talk about the ORM's such as (Hibernate, JPA), it follows following steps:
1. Create/Load the configuration
2. Build the Sessionfactory
3. Build the Session
4. Access the Service (like CRUD operation)
5. Close the Session

Also as you have mention, when the network is slow or the database latency is more, we can expect to see more time lag
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel

Reply

0 aloks0189
Certified Lead Developer
over 8 years ago

@chetany i think process history holds overall time taken for this instance to get complete, rather than a particular node execution time, please correct me if i am wrong.

Also can you please let us know, how exactly are trying to capture the start and time and the difference between start and end time after the completion of Write to Data Store Node execution.

Yes, i agree with you that, Appian has to follow lot many set of steps before performing CRUD operation on DB. As Appian internally uses ORM (JPA, Hibernate) and ORM internally uses JDBC, so as per the JDBC internal mechanism, it has to follow various steps in order to perform any Operation on Relational Tabular Data store, these are:

1. Obtain connection to DB
2. Get a JDBC Statement
3. Execute the Statement
4. Close the Connection

Now when we talk about the ORM's such as (Hibernate, JPA), it follows following steps:
1. Create/Load the configuration
2. Build the Sessionfactory
3. Build the Session
4. Access the Service (like CRUD operation)
5. Close the Session

Also as you have mention, when the network is slow or the database latency is more, we can expect to see more time lag
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel

Children

No Data