SESSION

AT&T's Migration of Billions of Events Processing From Hadoop

Accept Cookies to Play Video

OVERVIEW

EXPERIENCEIn Person
TYPEBreakout
TRACKData Engineering and Streaming
INDUSTRYMedia and Entertainment
TECHNOLOGIESApache Spark, ETL, Orchestration
SKILL LEVELIntermediate
DURATION40 min
DOWNLOAD SESSION SLIDES

AT&T's strategic migration of Hadoop MapReduce jobs to Databricks for one of their network applications has resulted in significant cost and time efficiencies, setting a benchmark in big data processing and analytics. This presentation outlines the comprehensive approach and industry best practices employed by AT&T in this migration. The transition involved converting Hadoop MapReduce jobs to Spark jobs, which are natively supported by the Databricks Platform, leading to improved performance and scalability. This move resulted in a substantial 30% reduction in compute costs and more than halved the execution time, thereby enhancing AT&T's operational efficiency and productivity. The successful migration exemplifies the transformative potential of cloud-native platforms and underlines the value of adopting industry best practices in big data management and processing.

SESSION SPEAKERS

Praveen Vemulapalli

/Director- Technology
AT&T

Akshay Sharma

/Senior Solutions Consultant
Databricks