YoVDO

Scaling the Distributed Actor Runtime - PARTISAN

Offered By: USENIX via YouTube

Tags

USENIX Annual Technical Conference Courses Erlang Courses Scalability Courses

Course Description

Overview

Explore the design and implementation of PARTISAN, an alternative runtime system for distributed actor applications, in this conference talk from USENIX ATC '19. Delve into how PARTISAN improves scalability and reduces latency through dynamic overlay selection, named channels, and affinitized parallelism. Learn about the limitations of current distributed actor systems and how PARTISAN addresses these issues. Examine experimental results demonstrating significant improvements in scalability, throughput, and latency compared to Distributed Erlang. Gain insights into the potential impact of PARTISAN on distributed systems development and performance optimization.

Syllabus

Intro
Distributed Actors
Programming Model
Executive Summary We're going to look at how we can improve distributed actor
Limitations: Scalability
Limitations: Latency
Partisan
Dynamic Overlay Selection
Named Channels
Affinitized Parallelism
Experiments
Evaluating Scalability Distributed advertisement counter
Increasing Scalability
Reducing Latency: Microbenchmarks
Increasing Throughput: Echo Service
Increasing Throughput: KVS Service
Takeaways Distributed actor systems limited by implementation assumptions
Questions?


Taught by

USENIX

Related Courses

Financial Sustainability: The Numbers side of Social Enterprise
+Acumen via NovoEd
Cloud Computing Concepts: Part 2
University of Illinois at Urbana-Champaign via Coursera
Developing Repeatable ModelsĀ® to Scale Your Impact
+Acumen via Independent
Managing Microsoft Windows Server Active Directory Domain Services
Microsoft via edX
Introduction aux conteneurs
Microsoft Virtual Academy via OpenClassrooms