Bert Latamore0
@IBM Program Director @Vikram_Murali live now on #theCUBE @DVellante @JohnWalls21 http://bit.ly/2h0MHC...
Bert Latamore
We use #Kubernetes. All these are tools used throughout the valley. We are dockerizing them @Vikram_Murali @IBM #theCUBE http://bit.ly/2h0MHC...
Bert Latamore
The microservices architecture, for example Tensorflow is one, we can just attach and use. @Vikram_Murali @IBM #theCUBE http://bit.ly/2h0MHC...
Bert Latamore
With the microservice architecture it is very easy for us to add new technologies that would have required complete rearchitecting with earlier tech. @Vikram_Murali @IBM #theCUBE http://bit.ly/2h0MHC...
Bert Latamore
As part of our partnership we set up co-engineering teams with #Hortonworks. Phase 2 + 3 will be deeper integration. @Vikram_Murali @IBM #theCUBE http://bit.ly/2h0MHC...
Bert Latamore
In 3rd phase we will use #Yarn for resource management. #Hortonworks is putting resources into Yarn. @Vikram_Murali @IBM #theCUBE http://bit.ly/2h0MHC...
Bert Latamore
We need knowledge on our side to know what the Hortonworks developers are doing. IBM Research has worked on a lot of the open source libraries. @Vikram_Murali @IBM #theCUBE http://bit.ly/2h0MHC...
Bert Latamore
#PixieDust is an example of a library that IBM Research developed and donated to Open Source. @Vikram_Murali @IBM #theCUBE http://bit.ly/2h0MHC...
Bert Latamore
The #1 problem for Data Scientists is getting clean data & access all the data. So much data out there. @Vikram_Murali @IBM #theCUBE http://bit.ly/2h0MHC...
Bert Latamore
Data scientists do not want to spend time cleaning up the data. @Vikram_Murali @IBM #theCUBE http://bit.ly/2h0MHC...
Bert Latamore
As part of our partnership with #Hortonworks we have connectors so the data scientists don't have to find the data. @Vikram_Murali @IBM #theCUBE http://bit.ly/2h0MHC...