[
https://issues.apache.org/jira/browse/FINERACT-1114?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Roshan Ratnaweera updated FINERACT-1114:
----------------------------------------
Description:
# Loan disbursement
# viewing the net amount of teller (Teller cash management)
# settle cash
When a user attempts to perform one of the above functions, a "select query" is
activated to retrieve the net amount of his cash on hand. It checks for every
transaction he made such as teller cash management, savings transactions, FD
transactions and loan transactions.
This is the reason why the performance of the server is greatly reduced after
using Mifos X for about two years. No matter how many server resources you
increase, slowing down is unavoidable.
If we can implement a day end process of updating the summary of each teller's
cash on hand to a separate table, it will be a solution for this. Then when you
try to do one of the above tasks, instead of checking all the transactions, you
can get the relevant teller's net amount very quickly by referring to that
table.
herewith attached slow queries also.
was:
Server configuration is *16 vCPU, 64GB RAM* and SSD storage. There are maximum
concurrent user count is 30 (from 15 branches)
# Loan disbursement
# viewing the net amount of teller (Teller cash management)
Above two tasks are taking a long time to process. When a bunch of these two
tasks are running, server CPU are 100% utilized by relevant sql select queries.
When the server is not going to normal status, I am killing frozen select
queries. Then suddenly the server is going back to normal.
I have noticed that these delayed two queries, read all transactions of
relevant users. I think, that is why it takes a long time to process. Are any
solutions available for this ?
> Slow Query
> ----------
>
> Key: FINERACT-1114
> URL: https://issues.apache.org/jira/browse/FINERACT-1114
> Project: Apache Fineract
> Issue Type: Bug
> Components: Performance
> Environment: Server configuration is 16 vCPU, 64GB RAM and 250GB SSD
> storage.
> Mifos setup has been done using AWS-AMI
> Reporter: Roshan Ratnaweera
> Priority: Critical
> Labels: scalability
> Attachments: Slow Query Loan Disbursement.sql, Slow Query Teller Cash
> Management Transactions.sql
>
>
>
> # Loan disbursement
> # viewing the net amount of teller (Teller cash management)
> # settle cash
> When a user attempts to perform one of the above functions, a "select query"
> is activated to retrieve the net amount of his cash on hand. It checks for
> every transaction he made such as teller cash management, savings
> transactions, FD transactions and loan transactions.
> This is the reason why the performance of the server is greatly reduced after
> using Mifos X for about two years. No matter how many server resources you
> increase, slowing down is unavoidable.
> If we can implement a day end process of updating the summary of each
> teller's cash on hand to a separate table, it will be a solution for this.
> Then when you try to do one of the above tasks, instead of checking all the
> transactions, you can get the relevant teller's net amount very quickly by
> referring to that table.
> herewith attached slow queries also.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
