Sorry-- thanks for pinging back on the thread.
The HealthVault PPE environment isn't necessarily built out for direct perf/load/stress testing, and direct stress could impact other partners, as this is a shared environment. However, PPE (and production) is scaled out to handle heavy volumes of traffic across many simultaneous partners, so it can handle all the load its seen so far, and we can easily scale it up more in the future as needs increase.
We encourage partners to test their own application/system to identify potential bottlenecks, and if that puts indirect load on the PPE environment, that's likely fine. Partners have done this in the past, and typically when partner apps are tested at appropriate loads the underlying load on the HealthVault services is relatively low.
What we want to avoid is partners running automated load tests directly against HealthVault itself, as that could impact other partners doing development work.
Our production environment is, of course, scaled up even more than PPE-- but we don't support direct perf testing there either, as that can have partner impact in a shared environment.
Please keep in mind that if you do put a significant (very large) amount of load on our services, it may appear as a DOS attack, and our ops team may start blocking connection requests.
If there is a strong need to to testing which will require putting a large load on the HealthVault services, please contact us at
hvbd@microsoft.com with as much information about the type and amount of load, and the rationale as to why the testing is necessary, and we'll work with you to find the best solution.
