On Tue, 1 Sept 2026, 23:15 Tofunmi Oguntibeju, <[email protected]>
wrote:

> Hi,
>
> I'd suggest Apache JMeter — mature, widely used, GUI/CLI and highly
> configurable (I have this setup in a cluster).
>
> I also came across k6 recently — I think it is a nice alternative based on
> google results.
>
> Critical flows worth testing:
>
>    - Client onboarding: POST /clients
>    - Loan lifecycle: submit POST /loans → ?command=approve →
>    ?command=disburse → repayme
>    - Savings: create → approve → activate → deposit/withdrawal
>    - Journal entries: POST /journalentries (DB-heavy writes)
>    - Login: POST /api/v1/authentication
>
> I think your goal should be correlating API latency with Infra values so
> start with default values and modify values as you ramp up.
>
> JVM heap, connection pool, CPU usage, memory are some values worth
> monitoring.
>
>
> *N.B:*
>
>    - Login isn't a business workload that is heavy per say but gives you
>    an idea of how long it takes to authenticate with Fineract
>    - In Prod, real clients authenticate once and reuse basic header for
>    subsequent calls so doing a lookup for each activity would pollute your 
> TPS.
>    - So login once per user in a thread group, then save and reuse the
>    returned values in Auth Headers
>
>
Thanks Tofunmi for the guidance. I appreciate it. I am running 8 vCPUs and
32 GB RAM hardware running Ubuntu Linux server 24.04 OS. Using Jmeter test
TPS does not cross beyond 68 TPS. Am I missing anything?

Please let me know if you need anything from me regarding load testing. I
look forward to hearing from you.

Best regards,

Kaushal

>
>
> On Tue, Sep 1, 2026 at 5:30 PM Kaushal Shriyan <[email protected]>
> wrote:
>
>> Hi,
>>
>> We are planning to conduct load testing of the Apache Fineract APIs to
>> evaluate the system's performance, stability, scalability, and response
>> times under different levels of concurrent users and API requests.
>>
>> The testing will cover key APIs and critical business transactions, with
>> a focus on:
>>
>> - API response time and throughput
>> - Concurrent request handling
>> - CPU and memory utilization
>> - Database performance
>> - Error rates and failed requests
>> - System behavior under normal and peak loads
>> - Identification of potential performance bottlenecks
>> - Overall system stability and scalability
>>
>> We will execute the tests in a controlled non-production environment
>> using representative test data and gradually increase the load to identify
>> the system's performance limits.
>>
>> The test results will be analyzed and documented, along with
>> recommendations for any required application, database, infrastructure, or
>> configuration-level optimizations.
>>
>> Please let us know if there are any specific APIs, transaction volumes,
>> concurrency levels, or performance benchmarks that should be included in
>> the testing scope.
>>
>> Additionally, please recommend any tools that can be used to perform load
>> testing of Apache Fineract APIs.
>>
>> Best regards,
>>
>> Kaushal
>>
>

Reply via email to