YoVDO

Memento: A Framework for Detectable Recoverability in Persistent Memory

Offered By: ACM SIGPLAN via YouTube

Tags

Persistent Memory Courses Thread Safety Courses

Course Description

Overview

Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore a groundbreaking framework for detectable recoverability in persistent memory through this 19-minute conference talk from PLDI 2023. Dive into Memento, a general and high-performance programming framework designed for detectably recoverable concurrent data structures in persistent memory. Learn how Memento supports primitive operations like checkpoint and compare-and-swap, and employs a timestamp-based recovery strategy for enhanced performance. Discover the framework's ability to ensure detectable recoverability during crashes while maintaining comparable performance to existing hand-tuned persistent data structures. Examine implementations of lock-free stack, list, queue, hash table, and combining queue that showcase Memento's capabilities in stress tests. Gain insights into the future of persistent memory technology and its potential impact on storage solutions combining DRAM performance with SSD durability.

Syllabus

[PLDI'23] Memento: A Framework for Detectable Recoverability in Persistent Memory


Taught by

ACM SIGPLAN

Related Courses

Create Your First Multithreaded Application in Java
Coursera Project Network via Coursera
Creational Design Patterns in Swift 5
Pluralsight
Advanced Threading in C#
LinkedIn Learning
How To Write Bulletproof Multi-Threaded C# Code
Udemy
Efficient Java Multithreading and Concurrency with Executors
Udemy