Hi guys
I'm trying to get my understanding clear about something. Not sure if this is the right place to post this,
so sorry if coming to the wrong place. This is maybe a silly question, but the point is lets say kibana is getting back data from the cluster at a high rate (say +1k events per sec) to visualiza it in real-time in the client-side, say a dashboard, and I need to display those events into a chart, say a line chart, but this is so much amount of events that the browser cannot mange it that and finally it will end up blocking the UI, even more when it's needed to parse those events in order to give the right format for the charting lib, kibana uses d3 I think with svg and DOM manipulation.
So in this scenario, what should be the right way to go? should those events be reduced at cluster side by elasticsearch, and send just few events to the client ? how kibana manages sampling strategies? should the cluster carry out all computations and free the browser from doing so ? or should it be managed by web workers at client side by kibana? ... is there a better approach out there? what do you thing it should be the better way for kibana to manage this @spalger?
thanks so much and again, apologies if going wrong with this question.
Hi guys
I'm trying to get my understanding clear about something. Not sure if this is the right place to post this,
so sorry if coming to the wrong place. This is maybe a silly question, but the point is lets say kibana is getting back data from the cluster at a high rate (say +1k events per sec) to visualiza it in real-time in the client-side, say a dashboard, and I need to display those events into a chart, say a line chart, but this is so much amount of events that the browser cannot mange it that and finally it will end up blocking the UI, even more when it's needed to parse those events in order to give the right format for the charting lib, kibana uses d3 I think with svg and DOM manipulation.
So in this scenario, what should be the right way to go? should those events be reduced at cluster side by elasticsearch, and send just few events to the client ? how kibana manages sampling strategies? should the cluster carry out all computations and free the browser from doing so ? or should it be managed by web workers at client side by kibana? ... is there a better approach out there? what do you thing it should be the better way for kibana to manage this @spalger?
thanks so much and again, apologies if going wrong with this question.