Load Testing Policy
Nanonets recognizes that customers may occasionally need to perform load tests against its production cloud service. In order to ensure a successful test and maintain a high quality of service for all customers, Nanonets has established the following guidelines. Any load testing in Nanonets must be conducted in accordance with this Policy.
Only customers who have purchased an Enterprise subscription may conduct load testing. Free, Developer, Pro and other non-Enterprise customers may not conduct load testing. Customers with an Enterprise subscription may request one load test (with up to 2 repeats) per year against an Nanonets production model. Performance and load testing is only allowed with Nnaonets' prior written approval. Once approved, testing can only target tenants that we have approved.
Nanonets reserves the right to reject the load test request or ask for modifications. Failure to abide by this policy may result in temporary blocking of access to a tenant until the issue is remediated.
You must file a load testing request by email to [email protected].
To be considered for approval, the request must:
Be filed at least 2 weeks prior to the desired test date; in many cases, Nanonets encourages one 1 month of advance notice to ensure time for a thorough review and any required modifications.
Be approved in writing before any testing is conducted.
Stay within our published production rate limits.
Include all information described below.
The load testing request must include the following:
A description of the test to be done
The name and region of the Nanonets model to be used during the test
The requested date and time of the test, including time zone
The requested duration of the test (2 hour maximum)
The platforms to be used for the test (desktop/laptop, iOS, Android, other)
The Nanonets features used during the test
The Nanonets API methods and endpoints to be used (for example
GET /api/v2/models)The maximum requests per second for each type of request or endpoint
The types of Nanonets connections involved in the test
The peak load, specified in requests-per-second, expected for each API endpoint or Nanonets feature involved in the test
An explanation/justification for the peak load numbers, including the size of the target user population and realistic estimates of logins per hour
The ramp-up rate for the test
Load testing windows are subject to availability so advance notice is highly recommended. Once approved, load testing windows will have a scheduled start and end time not to exceed two (2) hours in duration. All testing must begin and end during this window.
Nanonets strongly recommends including a brief "ramp up" period to the desired load test target numbers. For example, a load test request of 10 RPS might be preceded by three five minute periods: 5 minutes at 1 RPS, 5 minutes at 2 RPS, and 5 minutes at 5 RPS. This ramp up period allows Auth0 and the customer to observe and compare effects at increasing RPS levels prior to peak RPS. If a ramp up period is not possible, please indicate why.
