Overview
The course teaches how to design a fault-tolerant and scalable data store that combines causal and strong consistency to address the challenges of maintaining liveness in geographically distributed data centers. The learning outcomes include understanding the trade-offs between consistency levels, implementing causal and strong transactions, and evaluating the effectiveness of the UniStore system. The course is intended for software engineers, system architects, and researchers interested in distributed systems and data store technologies. The teaching method involves presenting theoretical concepts, discussing the design of UniStore, and evaluating its performance through benchmarks and real-world scenarios.
Syllabus
Intro
Geo-replication
Fundamental trade-offs
The hybrid alternative
A banking application
UniStore: causal baseline
UniStore: strong txs
US data UniStore: key challenge center crashes replicating to
Related work
UniStore: uniformity
UniStore: other features
Evaluation: results
Conclusion
Taught by
USENIX