FAST (File and Storage Technologies) Courses

USENIX

FAST '20 - An Empirical Guide to the Behavior and Use of Scalable Persistent Memory

Syllabus: Intro Outline Background: Optane in the Machine Background: Optane Internals Background: Optane Interleaving Basics: Our Approach Test Platform Basics: Latency Basics: Bandwidth Avoid small random accesses Lessons: Optane Buffer Size Use ntstor…

• YouTube
• Less than 1 hour of material
• On-Demand
• Free Online Course

USENIX

FAST '22 - DeepSketch: A New Machine Learning-Based Reference Search Technique...

Syllabus: Intro Executive Summary Data Reduction in Storage Systems Post-deduplication Delta Compression Combines three different data-reduction approaches Overview of Post-Deduplication Delta Compression Lossless Compression Key Challenge: Reference Sea…

• YouTube
• Less than 1 hour of material
• On-Demand
• Free Online Course

USENIX

FAST '22 - Operational Characteristics of SSDs in Enterprise Storage Systems: A Large-Scale Field...

Syllabus: Computer Science UNIVERSITY OF TORONTO Research Questions • Performance and lifespan of SSDs are affected by operational characteristics. Systems Description . We focus on NetApp's enterprise storage systems What are NAND Usage Rates in the fie…

• YouTube
• Less than 1 hour of material
• On-Demand
• Free Online Course

USENIX

FAST '22 - Removing Double-Logging with Passive Data Persistence in LSM-tree based Relational...

Syllabus: Introduction Background Log Key Idea Challenges Solution Conclusion

• YouTube
• Less than 1 hour of material
• On-Demand
• Free Online Course

USENIX

FAST '22 - Hydra : Resilient and Highly Available Remote Memory

Syllabus: Intro Remote Memory is Vulnerable Remote Memory Performs Great! Performance Tradeoff for Resiliency Remote Memory w/ Hydra Data Plane Hydra Datapath - Asynchronously Encoded Write Hydra Datapath – Late-binding Read Replication-like Performance…

• YouTube
• Less than 1 hour of material
• On-Demand
• Free Online Course

USENIX

FAST '22 - FusionFS: Fusing I/O Operations using CISCOps in Firmware File Systems

Syllabus: Intro Storage Hardware and Software Trend. Evolving Storage with Fast Compute State-of-the-art Designs Common I/O Sequences in Application Storage Approaches Summary Compute Our Solution: Fusion FS Everlasting Debate FusionFS: RISC vs CISC oper…

• YouTube
• Less than 1 hour of material
• On-Demand
• Free Online Course

USENIX

FAST '22 - InfiniFS: An Efficient Metadata Service for Large-Scale Distributed Filesystems

Syllabus: Intro Outline Large-Scale Distributed Filesystem Filesystem Metadata Challenges of efficient Metadata Service InfiniFS Architecture Key Designs 1. Access-Content Decoupled Partitioning 2. Speculative Path Resolution Scalability Breakdown Conclu…

• YouTube
• Less than 1 hour of material
• On-Demand
• Free Online Course

USENIX

FAST '22 - ScaleXFS: Getting scalability of XFS back on the ring

Syllabus: Intro Hardware Trend Journaling in Filesystem Scalability Failure in XFS Journaling in Ext4 vs. in XFS Ext4 Scalable Filesystem Design of Ext4 Latency of metadata operation Lock contention in in-memory logging Locks in In-memory Logging Content…

• YouTube
• Less than 1 hour of material
• On-Demand
• Free Online Course

USENIX

FAST '22 - Practicably Boosting the Processing Performance of BFS-like Algorithms...

Syllabus: Intro Semi-External Graph System BFS-like Algorithms on Semi-External System Existing Work for Optimizing BFS-like Algorithms I/O Efficiency Motivation 1/O-Efficient Graph Ordering (IOE-Order) Out-Degree Binning (Conti.) Evaluation Setup Overal…

• YouTube
• Less than 1 hour of material
• On-Demand
• Free Online Course

USENIX

FAST '22 - PAIO: General, Portable I/O Optimizations With Minor Application Modifications

Syllabus: Intro Data-centric systems Challenge #1 PAIO design 1/0 differentiation Control plane interaction Tail Latency Control in LSM-based Key-Value Stores Mixture workload Read-heavy workload Write-heavy workload Paper

• YouTube
• Less than 1 hour of material
• On-Demand
• Free Online Course

USENIX

FAST '22 - FORD: Fast One-sided RDMA-based Distributed Transactions for Disaggregated Persistent...

Syllabus: Intro Disaggregated Persistent Memory Memory disaggregation RDMA-based Distributed Transaction Transaction on Disaggregated PM State-of-the-art Challenge 1 Overview Hitchhiked Locking Read and lock the read-write (RW) set in execution . Avoid s…

• YouTube
• Less than 1 hour of material
• On-Demand
• Free Online Course

USENIX

FAST '22 - NyxCache: Flexible and Efficient Multi-tenant Persistent Memory Caching

Syllabus: Introduction Memorybased today Inmemory keyword caches PMM throughput Cloud resource limiting approach Design new mechanism DRM interface analysis Demonstration Summary

• YouTube
• Less than 1 hour of material
• On-Demand
• Free Online Course

USENIX

FAST '22 - ctFS: Replacing File Indexing with Hardware Memory Translation through Contiguous File...

Syllabus: Introduction Precision Memory Indexing Design Challenges Hierarchy Code Layout Partition Layout Partition Organization Pswap Efficiency Performance Benchmarks

• YouTube
• Less than 1 hour of material
• On-Demand
• Free Online Course

USENIX

FAST '22 - HTMFS: Strong Consistency Comes for Free with Hardware Transactional Memory...

Syllabus: Intro Performance vs Consistency Strong consistency Persistent Memory Hardware Transactional Memory Crash Consistency Mechanism Compari Challenges HOP Hardware-assisted Optimistic Persisten Use HOP to build HTMFS Data Accesses Atomic Memory All…

• YouTube
• Less than 1 hour of material
• On-Demand
• Free Online Course

USENIX

FAST '13 - Unioning of the Buffer Cache and Journaling Layers with Non-volatile Memory

Syllabus: Introduction Outline Example Why this happened Nonvolatile Memory Data Consistency Consistency EBJ Update Cache Performance System Crashes Experimental Results Conclusion Questions

• YouTube
• Less than 1 hour of material
• On-Demand
• Free Online Course