"Perhaps some built in visualization widget that can poll a backend for updates?"
Surely, but then someone need to code it. And a pre-built solution is never really as flexible as a customized dev. I don't see how we can create a generic component. Unless we decouple the data receiver component from the display component and make them pluggable. On Wed, Aug 24, 2016 at 5:51 PM, Vinay Shukla <vinayshu...@gmail.com> wrote: > Is there a way to make it easier for our Zeppelin users? Perhaps some > built in visualization widget that can poll a backend for updates? > > > On Wednesday, August 24, 2016, DuyHai Doan <doanduy...@gmail.com> wrote: > >> "I want the updates to be pushed to the dashboard I dont want keep >> running the query over and over again or do some sort of polling." >> >> In this case the only solution for you is: >> >> 1) use %angular interpreter >> 2) import a JS library in a paragraph and do some JS coding to open a >> web-socket channel to the data source so that the new updates can be pushed >> directly to the paragraph >> >> >> >> On Wed, Aug 24, 2016 at 1:15 PM, kant kodali <kanth...@gmail.com> wrote: >> >>> Hi, >>> >>> Sure, I can either write it to HDFS or Cassandra but not HBASE since we >>> don't use it but how about constant updates? you know what I mean? I want >>> the updates to be pushed to the dashboard I dont want keep running the >>> query over and over again or do some sort of polling. >>> >>> Thanks! >>> >>> >>> >>> On Wed, Aug 24, 2016 3:36 AM, Vinay Shukla vinayshu...@gmail.com wrote: >>> >>>> Hey Kant, >>>> >>>> For fast data updates via streaming the normal flow is to put the >>>> aggregated results in things like HBase. You could then use Zeppelin's >>>> Phoenix interpreter to read that data. >>>> Not sure if this fits your requirements. >>>> >>>> Thanks, >>>> Vinay >>>> >>>> On Wednesday, August 24, 2016, kant kodali <kanth...@gmail.com> wrote: >>>> >>>> Hi Guys, >>>> >>>> is there way to update(push latest data to but not pull) Zeppelin >>>> dashboard ? for example I get bunch of data to my spark streaming cluster >>>> and as I get more and more data I want to compute and write it somewhere >>>> such that my Zeppelin dashboard is updated? If so, where would be the best >>>> place to write the latest results? >>>> >>>> Thanks, >>>> Kant >>>> >>>> >>
