Distributed systems are increasingly shifting from centralized to decentralized architectures.
Decentralized systems replicate data to improve the system’s scalability and availability. However, the CAP theorem makes it impossible for partition-tolerant systems to be both strongly consistent and continuously available.
Hence, modern applications tend to relax the consistency guarantees when possible and adhere to weaker consistency models which guarantee better availability and performance.
Building highly available systems currently remains cumbersome because replicas can be modified concurrently. Hence, programmers need to manually deal with inconsistent states. Solving such inconsistencies requires advanced knowledge of conflict resolution techniques which can only be expected from experts in distributed systems.
The goal of my research is to design and implement a general-purpose framework for highly available distributed systems. My framework will provide programmers with the means to trade off the availability and consistency of individual objects without requiring all objects to adhere to the same consistency model. My framework will also empower developers to control the dissemination of replicated data. To achieve this goal three concrete objectives will be completed:
2018 - Present
2016 - 2018
Software Languages and Software Engineering
Honors: summa cum laude
2016 - 2017
Theme: Artificial Intelligence, Cognition and Creativity
2013 - 2016
Honors: magna cum laude
March 2017 to present
Senior Web Developer
Santa monica, Los angeles
March 2017 to present
Senior Web Developer
Santa monica, Los angeles
March 2017 to present
Senior Web Developer
Santa monica, Los angeles