You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@fineract.apache.org by "Avik Ganguly (Jira)" <ji...@apache.org> on 2020/05/25 13:47:00 UTC

[jira] [Commented] (FINERACT-427) Paginate input queries for jobs.

    [ https://issues.apache.org/jira/browse/FINERACT-427?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17116030#comment-17116030 ] 

Avik Ganguly commented on FINERACT-427:
---------------------------------------

[~kumaranath_fernando] : I remember there were some challenges for some of the jobs. In the end, which of the above 6 jobs were you able to paginate in the merged PR?

> Paginate input queries for jobs.
> --------------------------------
>
>                 Key: FINERACT-427
>                 URL: https://issues.apache.org/jira/browse/FINERACT-427
>             Project: Apache Fineract
>          Issue Type: Improvement
>            Reporter: Avik Ganguly
>            Priority: Minor
>              Labels: gsoc, p2, performance
>
> Paginate read queries for following jobs :
> Add Accrual Transactions
> Add Periodic Accrual Transactions
> Add Accrual Transactions For Loans With Income Posted As Transactions
> Generate Loan Loss Provisioning
> Post Interest for Savings
> Update Loan Summary
> Add arbitrary page size (32000) for now which can be later replaced by the product of two configurations (batch size * thread count for that executor service pool) when jobs are made concurrent. 
> If paginating the query outside can split a loan across pages, paginate the primary entity, i.e. loans or savings. 
> Ex :- If accrual read query will fetch 3 rows for same loan (3 repayment schedule entries for which accrual is pending) and page size is 2,  loan will get split across 2 pages. Avoid this by paginating the loan entity inside the query.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)