You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Have the hub (upon api invocation) send a ping to all modules, including the time sent. Each module responds with both the incoming timestamp and the time it received the ping. The data will let us estimate queue-length and load related latency factors.
Yeah, let's talk about this when it's convenient.
The text was updated successfully, but these errors were encountered:
This seems reasonable to implement as a command implemented by the manager
class.
What interface do you imagining using to expose that functionality, or
would it be reasonable for it to run periodically and print out to the logs?
On May 7, 2015 3:40 PM, "Molly Ling" [email protected] wrote:
Have the hub (upon api invocation) send a ping to all modules, including
the time sent. Each module responds with both the incoming timestamp and
the time it received the ping. The data will let us estimate queue-length
and load related latency factors.
Yeah, let's talk about this when it's convenient.
—
Reply to this email directly or view it on GitHub #268.
A promise that fires at the specified rate. It's parameters are the list of ping times - one entry in the list corresponds to reach module, containing the three times of the ping - hub transmit, module reply, and hub receive.
E.g. a list of objects, where NNN represents the value from Performance.Now() at each point:
[{pingSent: NNN, pongSent: NNN, pongRecieved: NNN, module: "logger"}, {....}]
Have the hub (upon api invocation) send a ping to all modules, including the time sent. Each module responds with both the incoming timestamp and the time it received the ping. The data will let us estimate queue-length and load related latency factors.
Yeah, let's talk about this when it's convenient.
The text was updated successfully, but these errors were encountered: