Virtual Users, VUs, are the simulated clients we load onto your system. They load objects in parallel, just like a real browser would. To know how many you need for testing, apply this formula: VUs = (peak hourly_visits * avg. visit_duration in seconds) / 3600.
Requests Per Second, RPS, measures the estimated throughput of load we will generate onto your system. RPS is an estimated calculation based on:
- Each virtual user is capable of generating 25 concurrent TCP connections, when scripted using `http.request_batch`.
- An average request response time of 500ms.
Applying these two constants yields: RPS/VU = 25*(1000/500) = 50.
To determine the Load Impact account level you need, estimate your throughput needs and find the subscription that best matches. To reverse the estimates, divide your target RPS by 50 (or whatever the formula yields with your realistic system response times) to yield the VU levels you need. See this article for more details on how to set up your throughput in Load Impact.
Yes. If you need any help at all with testing and/or results analysis, we encourage you to contact us to receive a quote for our professional services.
We offer the option to install Server Metrics Agents on your target machines. This makes it possible to collect information like, CPU usage, memory usage, disk I/O, network I/O, during testing. You can also add other custom metrics using Nagios plugins.
- VUs: Virtual Users: simulated virtual users or clients that concurrently generate load onto your system, given one or many scripted user scenarios.
- RPS: Requests Per Second: a measure of throughput as number of requests generated onto a system every second. See previous question for details on how to convert VUs to RPS.
- Clients/Test: An accumulated number of requests (or scenarios) generated throughout the runtime of a single test. Load Impact does not currently support this way of measuring scale and sizing of a test. We find the VU and RPS approaches are best.
Yes, you can stop your monthly subscription anytime you want. You will not be charged beyond the month you decide to cancel.
Yes, you can downgrade or upgrade your subscription at anytime. You can also hold two subscriptions simultaneously. But if you wish to change plans, please make sure to cancel the first one before purchasing the other.
Yes. Load impact is a Subscription based service. Users are able to cancel their subscription at any time during their term. Service will be retained at the subscribed level until the last day of the subscription. To check your current expiration/renewal date, please check the billing menu in app. Load Impact does not offer refunds on subscriptions.
We offer variable rate subscriptions that allow you to run as many individual tests as you want - up to a certain load level. As your business grows, we'll automatically move you up to a higher load testing level. We can also create custom plans to match your test frequency and load volume needs. Please contact us for a quote.
Yes, we do offer a significant discount for non-profit organizations. Contact us with a description of your needs and we will be happy to oblige.
Yes. Any unused test credits will rollover month to month.
Yes, if the test was aborted by the system you will get a full refund. If the test was aborted by you, then you will get a partial refund depending on how early you abort. But we're flexible, so contact us if you would like us to honor a full refund request. This policy applies to pay-per-test credits only.