Karthik Ramasamy's presentation was part of a panel discussion on Stream Processing Systems on January 20th, 2016 led by Ben Lorica (O'Reilly Media) with panelists: Jay Kreps (Confluent), M.C. Srivas (MapR), Nikita Shamgunov (MemSQL), Ram Sriharsha (Hortonworks)
2. 2
Value of Real Time DataIt’s contextual
[1] Courtesy Michael Franklin, BIRTE, 2015.
3. 3
Heron
Batching of tuples
Amortizing the cost of transferring tuples
Task isolation
Ease of debug-
ability/isolation/profiling
Fully API compatible with Storm
Directed acyclic graph
Topologies, Spouts and Bolts
Support for back pressure
Topologies should self adjusting
gUse of main stream languages
C++, Java and Python
Efficiency
Reduce resource consumptionG
Design: Goals
7. Auto scaling the system in the presence of unpredictability
7
Technology Challenges
The Road Ahead
Auto tuning of real time analytics jobs/queries
Exploiting faster networks for efficiently moving data
Ä
Ü
J