Yuanhao Sun, Transwarp Technology | Big Data New York City 2016
summary
Transwarp claims a Hadoop/Spark platform deployed by 500 customers that pairs a SQL engine (about 5 million lines of code) with a streaming engine built atop a modified Spark core to be event driven. Compatibility with Oracle/DB2 PL SQL reaches roughly 90%, supports transactions on HDFS, and ingests up to 10 million inputs per second for wind-farm IoT use cases. Deployments use ~100-node edge clusters, Kafka connectors, optional local Pi databases, and achieve latencies from several to tens of milliseconds.
AI-written summary · about
chapters (12)
- 0:00Start
- 0:2001. Yuanhao Sun, Transwarp Technology, Visits theCUBE !.
- 0:4902. What Are Your Impressions Of The Show.
- 1:1303. Tell Us About Your Petigree In China.
- 2:5904. How Much Effort Was It For You To Keep Compatability.
- 4:5505. What Is ETL Offlload Typically Done On.
- 6:5406. If You Build On Spark How Do You Compensate For Spark Not Having Concurrency.
- 8:5607. What Does Your Product Set Look Like In Internet Of Things.
- 11:2708. Who Are Some Of The Early Customers Taking Advantage Of Streaming.
- 13:5609. How Much Of The System Is Operating On The Mountain On The Wind Farm.
- 15:2010. What Type Of Analysis Are You Doing Near The Edge.
- 16:2511. How Did You Do All This In Three Years.
