Jira |
---|
server | ONAP Jira |
---|
serverId | 425b2b0a-557c-3c0c-b515-579789cceedb |
---|
key | CPS-1172 |
---|
|
# | Issue | Notes | Decision |
---|
1 | Agree registration performance KPI | - Size of request (e.g. 100 cm handles, total size <10Kb)
- Concurrent requests for registration (e.g. 4)
- Request frequency (e.g. 4 concurrent request each 5 seconds)
- Response time of each request (e.g. <4sec)
- Single CPS instance
- No other workload (module sync 'delayed')
- Default resources as defined in our helm charts
|
|
2 | Agree test environment (talk with Andrew Fenner) | CI Job(s) required? |
|
3 | Agree technology used to test | - Groovy/Spock
- CSIT
- Python
- K6 (not recommended)
- Other script using curl?
|
|
4 | Additional requirements to be able to run test | 'dummy' DMI PLugin for testing ? (possibly not needed yet for this testcase) |
|
TBD
Here the various technologies will be assessed to determine which can offer the most flexibility, maintainability, and reliability.
There is an existing Groovy/Spock performance test located in org.onap.cps.ri.performance.CpsToDataNodePerfSpec that could be used as the basis for this...
This test
# | Pros | Cons | Notes |
---|
1 | - A lot of knowledge in the team
- Easier to control
| - Might not be at the correct loacation - too low down?
- Adds additional time to tests during clean install
| - Is it too low down in the stack
|
# | Pros | Cons | Notes |
---|
1 | - A lot of knowledge in the team
- Easier to control
| - Adds additional time to tests during clean install
|
|
# | Pros | Cons | Notes |
---|
1 |
| - There is a lack of knowlesge in the team and ramp up would e necessary
|
|
Wwe can run a script as part of CSIT.. could itbe done in maven also?
# | Outcome/AC | Notes | Done |
---|
1 | Document proposal on Wiki and share with team and stakeholder |
|
|
2 | Agree with team and stakeholders on options proposed in Spike |
|
|