Upload
temple
View
0
Download
0
Embed Size (px)
Citation preview
Solving the Scalability Challenge from the Ground Up
Justin Y. Shi | [email protected]
Data Cloud 20156th International Workshop on Data Intensive Computing in the Clouds
Lessons from the History
Circuit‐Switching NetworkPacket‐Switching Network
• Fast• Expensive to Maintain• Difficult to Scale• (Dedicated switch to data at a time)
• Slower• Cheap to Maintain• Infinitely Scalable• (Data decoupled from
switches)
TodayAll Internet Traffic are Packet Switched
The Scalability Dilemma is …
“the Dedicated Resource Syndrome” Fixed program/data ‐processor binding Decoupled ‐>Unbounded Growth
Reliability Performance
Correctness of Distributed Program‐ProgramCoordination – One Way to Look at This Mess…
Program and Data Must be Decoupled from Hardware
Requirement 1: Zero Data Loss
Requirement 2: Zero Single Point Failure
Requirement 3: Infinitely Scalable
Summary
CAP Should All be Satisfied
Zero Single Point Failure
Decoupling is the Only SolutionVisit Booth #299 for More Details
Acknowledgements
• Reported CI architecture research is supported in part by National Science Foundation (MRI Grant #CNS0958854)
• DI architecture research is supported in part by Ben Franklin Technology Partners and private investors of Parallel Computers Technology Inc.
• NCAR for Yellowstone benchmark effort