Known Knowns and Unknowns Near-realtime Earth Observation Via Query Bifurcation in Serval
Burstable Cloud Block Storage with Data Processing Units
OmniCache Collaborative Caching for Near-storage Accelerators
Ethane An Asymmetric File System for Disaggregated Persistent Memory
Seer Enabling Future-Aware Online Caching in Networked Systems
Scalable Billion-point Approximate Nearest Neighbor Search Using SmartSSDs
Optimizing Resource Allocation in Hyperscale Datacenters Scalability, Usability, and Experiences
μSlope High Compression and Fast Search on Semi-Structured Logs
Parrot Efficient Serving of LLM-based Applications with Semantic Variable
Sabre Hardware-Accelerated Snapshot Compression for Serverless MicroVMs
Removing Obstacles before Breaking Through the Memory Wall A Close Look at HBM Errors in the Field
An Empirical Study of Rust-for-Linux The Success, Dissatisfaction, and Compromise
FIFO Queues are all You Need for Cache Eviction
ATC '24 经历分享
Characterization of Large Language Model Development in the Datacenter
In-Memory Key-Value Store Live Migration with NetMigrate
StreamCache Revisiting Page Cache for File Scanning on Fast Storage Devices
MinFlow High-performance and Cost-efficient Data Passing for I/O-intensive Stateful Serverless Analytics
VBase Unifying Online Vector Similarity Search and Relational Queries via Relaxed Monotonicity
Albis High-Performance File Format for Big Data Systems
ServerlessLLM Locality-Enhanced Serverless Inference for Large Language Models
TRINITY A Fast Compressed Multi-attribute Data Store
A Quantitative Analysis and Guidelines of Data Streaming Accelerator in Modern Intel Xeon Scalable Processors
We Ain’t Afraid of No File Fragmentation Causes and Prevention of Its Performance Impact on Modern Flash SSDs
MiDAS Minimizing Write Amplification in Log-Structured Systems through Adaptive Group Number and Size Configuration
Symbiosis The Art of Application and Kernel Cache Cooperation
What’s the Story in EBS Glory Evolutions and Lessons in Building Cloud Block Store
Gemini Fast Failure Recovery in Distributed Training with In-Memory Checkpoints
FAST 24 论文速览
ELECT Enabling Erasure Coding Tiering for LSM-tree-based Storage
An Empirical Evaluation of Columnar Storage Formats
BTRBLOCKS Efficient Columnar Compression for Data Lakes
Enabling High-Performance and Secure Userspace NVM File Systems with the TRIO Architecture
RON One-Way Circular Shortest Routing to Achieve Efficient and Bounded-waiting Spinlocks
Nodens Enabling Resource Efficient and Fast QoS Recovery of Dynamic Microservice Applications in Datacenters
A Cloud-Scale Characterization of Remote Procedure Calls
Calcspar A Contract-Aware LSM Store for Cloud Storage with Low Latency Spikes
Elf Erasing-based Lossless Floating-Point Compression
XFaaS Hyperscale and Low Cost Serverless Functions at Meta
Take Out the TraChe Maximizing (Tra)nsactional Ca(che) Hit Rate
Design Considerations and Analysis of Multi-Level Erasure Coding in Large-Scale Data Centers
Language Model is Compression
eZNS An Elastic Zoned Namespace for Commodity ZNS SSDs
TreeSLS A Whole-system Persistent Microkernel with Tree-structured State Checkpoint on NVM
SOSP 23 论文速览
Ensō A Streaming Interface for NIC-Application Communication
Efficient Memory Management for Large Language Model Serving with PagedAttention
Light-Dedup A Light-weight Inline Deduplication Framework for Non-Volatile Memory File Systems
Zebra ZRWA-EnaBled Redundant Array of Zoned Namespace SSDs
LLFree Scalable and Optionally-Persistent Page-Frame Allocation
Explore Data Placement Algorithm for Balanced Recovery Load Distribution
面向新型计算应用构建存储系统的若干思考
TiDedup A New Distributed Deduplication Architecture for Ceph
MedCompressor：医学图像压缩
Chimp Efficient Lossless Floating Point Compression for Time Series Databases
Selection Pushdown in Colum Stores using Bit Manipulation Instructions
Fisc A Large-scale Cloud-native-oriented File System
Orca A Distributed Serving System for Transformer-Based Generative Models
Hyrax Fail-in-Place Server Operation in Cloud Platforms
Multi-view Feature-based SSD Failure Prediction What, When, and Why
Unlocking unallocated cloud capacity for long, uninterruptible workloads
SMRStore A Storage Engine for Cloud Object Storage on HM-SMR Drive
HadaFS A File System Bridging the Local and Shared Burst Buffer for Exascale Supercomputers
Shard Manager A Generic Shard Management Framework for Geo-distributed Applications
CFS Scaling Metadata Service for Distributed File System via Pruned Scope of Critical Sections
Time Series Data Encoding for Efficient Storage A Comparative Analysis in Apache IoTDB
Eurosys 23 论文速递
InftyDedup Scalable and Cost-Effective Cloud Tiering with Deduplication
Integrated Host-SSD Mapping Table Management for Improving User Experience of Smartphones
Cachew ML input Data Processing as a Service
SDC Study
RAIZN Redundant Array of Independent Zoned Namespaces
Perseus A Fail-Slow Detection Framework for Cloud Storage Systems
GL-Cache Group-level learning for efficient and high-performance caching
ADOC Automatically Harmonizing Dataflow Between Components in Log-Structured Key-Value Stores for Improved Performance
ParaRC Embracing Sub-Packetization for Repair Parallelization in MSR-Coded Storage
Hi-Speed DNN Training with Espresso Unleashing the Full Potential of Gradient Compression with Near-Optimal Usage Strategies
TencentCLS The Cloud Log Service with High Query Performances
StRAID Stripe-threaded Architecture for Parity-based RAIDs with Ultra-fast SSDs
Hubble Performance Debugging with In-Production, Just-In-Time Method Tracing on Android
Looking Beyond GPUs for DNN Scheduling on Multi-Tenant Clusters
Demystifying and Checking Silent Semantic Violations in Large Distributed Systems
LogGrep+ improve retrieval efficiency on highly compressed cloud log with encoding aware schemes
ResPCT Fast Checkpointing in Non-volatile Memory for Multi-threaded Applications
Carbink Fault-Tolerant Far Memory
XRP In-Kernel Storage Functions with eBPF
Owl Scale and Flexibility in Distribution of Hot Content
Tiger Disk-Adaptive Redundancy Without Placement Restrictions
Neural Compression Review
CompressDB Enabling Efficient Compressed Data Direct Processing for Various Databases
Hydra Resilient and Highly Available Remote Memory
LogGrep
宽列纠删码
Zeus Locality-aware Distributed Transactions
Metastable Failures in the Wild
Pacman An Efficient Compaction Approach for Log-Structured Key-Value Store on Persistent Memory
Direct Access, High-Performance Memory Disaggregation with DirectCXL
BlockFlex Enabling Storage Harvesting with Software-Defined Flash in Modern Cloud Platforms
TriCache A User-Transparent Block Enabling High-Performance Out-of-Core Processing with In-Memory Programs
Shard Manager A Generic Shard Management Framework for Geo-distributed Applications
Geometric Partitioning Explore the Boundary of Optimal Erasure Code Repair
Microsecond-scale Preemption for Concurrent GPU-accelerated DNN Inferences
MorphStore Analytical Query Engine with a Holistic Compression-Enabled Processing Model
NVMe SSD Failures in the Field the Fail-Stop and the Fail-Slow
ctFS Replacing File Indexing with Hardware Memory Translation through Contiguous File Allocation for Persistent Memory
Good to the Last Bit Data-Driven Encoding with CodecDB
Where did my 256 GB go? A Measurement Analysis of Storage Consumption on Smart Mobile Devices
Adaptive Compression for Fast scans on string columns
p2KVS a Portable 2-Dimensional Parallelizing Framework to Improve Scalability of Key-value Stores on SSDs
存储盘用户画像刻画
DeltaFS A Scalable No-Ground-Truth Filesystem For Massively-Parallel Computing
Operational Characteristics of SSDs in Enterprise Storage Systems A Large-Scale Field Study
DEPART Replica Decoupling for Distributed Key-Value Storage
TVStore Automatically Bounding Time Series Storage via Time-Varying Compression
JSON Tiles Fast Analytics on Semi-Structured Data
Accelerating XOR-based erasure coding using program optimization techniques
DeepSketch A New Machine Learning-Based Reference Search Technique for Post-Deduplication Delta Compression
Improving the Reliability of Next Generation SSDs using WOM-v Codes
InfiniFS An Efficient Metadata Service for Large-Scale Distributed Filesystems
iVPF Numerical Invertible Volume Preserving Flow for Efficient Lossless Compression
The what, The from, and The to The Migration Games in Deduplicated Systems
RDMA is Turing complete, we just did not know it yet!
Closing the B+ -tree vs LSM-tree Write Amplification Gap on Modern Storage Hardware with Built-in Transparent Compression
FaaSNet Scalable and Fast Provisioning of Custom Serverless Container Runtimes at Alibaba Cloud Function Compute
LogECMem Coupling Erasure-Coded In-Memory Key-Value Stores with Parity Logging
Modernizing File System through In-Storage Indexing
Witcher Systematic Crash Consistency Testing for Non-Volatile Memory Key-Value Stores
Tsunami A Learned Multi-dimensional Index for Correlated Data and Skewed Workloads
VF2Boost Very Fast Vertical Federated Gradient Boosting for Cross-Enterprise Learning
WineFS a hugepage-aware file system for persistent memory that ages gracefully
Tiny-Tail Flash Near-Perfect Elimination of Garbage Collection Tail Latencies in NAND SSDs
KVIMR Key-Value Store Aware Data Management Middleware for Interlaced Magnetic Recording Based Hard Disk Drive
What is DNA Storage
IODA A Host/Device Co-Design for Strong Predictability Contract on Modern Flash Storage
D2FQ Device-Direct Fair Queueing for NVMe SSDs
Scaling Large Production Clusters with Partitioned Synchronization
FragPicker A New Defragmentation Tool for Modern Storage Devices
ZNS+ Advanced Zoned Namespace Interface for Supporting In-Storage Zone Compaction
Characterizing and Optimizing Remote Persistent Memory with RDMA and NVM
Kangaroo Caching Billions of Tiny Objects on Flash
FlashNeuron SSD-Enabled Large-Batch Training of Very Deep Neural Networks
Exploring the Design Space of Page Management for Multi-Tiered Memory Systems
NanoLog A Nanosecond Scale Logging System
OSCA An Online-Model Based Cache Allocation Scheme in Cloud Block Storage Systems
Optimizing Storage Performance with Calibrated Interrupts
Aurogon Taming Aborts in All Phases for Distributed In-Memory Transactions
Differentiated Key-Value Storage Management for Balanced I/O Performance
Boosting Full-Node Repair in Erasure-Coded Storage
Privacy Budget Scheduling
Jump-Starting Multivariate Time Series Anomaly Detection for Online Service Systems
Bridging the Performance Gap for Copy-based Garbage Collectors atop Non-Volatile Memory
CLP Efficient and Scalable Search on Compressed Text Logs
Oort Efficient Federated Learning via Guided Participant Selection
Rearchitecting Linux Storage Stack for µs Latency and High Throughput
Clay Codes Moulding MDS Codes to Yield an MSR Code
ParaFS A Log-Structured File System to Exploit
降低宽列纠删码的长尾延迟
轻量级神经网络超参数调优并行训练框架
宽列纠删码存储的设计与实现
基于超算的RSA分解因子平台搭建
Achieving Low Tail-latency and High Scalability for Serializable Transactions in Edge Computing
RackSched A Microsecond-Scale Scheduler for Rack-Scale Computers
ROART Range-query Optimized Persistent ART
REMIX Efficient Range Query for LSM-trees
Gandiva Introspective Cluster Scheduling for Deep Learning
CheckFreq Frequent, Fine-Grained DNN Checkpointing
Rethinking File Mapping for Persistent Memory
PIDS Attribute Decomposition for Improved Compression and Query Performance in Columnar Storage
Near-Optimal Latency Versus Cost Tradeoffs in Geo-Distributed Storage
The Storage Hierarchy is Not a Hierarchy Optimizing Caching on Modern Storage Devices with Orthus
Facebook’s Tectonic Filesystem Efficiency from Exascale
Exploiting Combined Locality for Wide-Stripe Erasure Coding in Distributed Storage
Concordia Distributed Shared Memory with In-Network Cache Coherence
SpanDB A Fast, Cost-Effective LSM-tree Based KV Store on Hybrid Storage
Cerebro A Data System for Optimized Deep Learning Model Selection
Analyzing and Mitigating Data Stalls in DNN Training
Microsecond Consensus for Microsecond Applications
How to Copy Files
DistCache Provable Load Balancing for LargeScale Storage Systems with Distributed Caching
FIRM An Intelligent Fine-grained Resource Management Framework for SLO-Oriented Microservices
PACEMAKER Avoiding HeART attacks in storage clusters with disk-adaptive redundancy
LinnOS Predictability on Unpredictable Flash Storage with a Light Neural Network
Sundial Fault-tolerant Clock Synchronization for Datacenters
From WiscKey to Bourbon- A Learned Index for Log-Structured Merge Trees
HetPipe Enabling Large DNN Training on (Whimpy) Heterogeneous GPU Clusters through Integration of Pipelined Model Parallelism and Data Parallelism
Fast RDMA-based Ordered Key-Value Store using Remote Learned Cache
OpenEC Toward Unified and Configurable Erasure Coding Management in Distributed Storage Systems
A Unified Architecture for Accelerating Distributed DNN Training in Heterogeneous GPU/CPU Clusters
Austere Flash Caching with Deduplication and Compression
An Empirical Guide to the Behavior and Use of Scalable Persistent Memory
BASTION A Security Enforcement Network Stack for Container Networks
GIFT A Coupon Based Throttle-and-Reward Mechanism for Fair and Efficient I/O Bandwidth Management on Parallel Storage Systems
MatrixKV Reducing Write Stalls and Write Amplification in LSM-tree Based KV Stores with a Matrix Container in NVM
A Deep Dive into DNS Query Failures
Ethane An Asymmetric File System for Disaggregated Persistent Memory
Scalable Billion-point Approximate Nearest Neighbor Search Using SmartSSDs
Removing Obstacles before Breaking Through the Memory Wall A Close Look at HBM Errors in the Field
An Empirical Study of Rust-for-Linux The Success, Dissatisfaction, and Compromise
ATC '24 经历分享
StreamCache Revisiting Page Cache for File Scanning on Fast Storage Devices
Albis High-Performance File Format for Big Data Systems
Nodens Enabling Resource Efficient and Fast QoS Recovery of Dynamic Microservice Applications in Datacenters
Calcspar A Contract-Aware LSM Store for Cloud Storage with Low Latency Spikes
Light-Dedup A Light-weight Inline Deduplication Framework for Non-Volatile Memory File Systems
LLFree Scalable and Optionally-Persistent Page-Frame Allocation
Explore Data Placement Algorithm for Balanced Recovery Load Distribution
TiDedup A New Distributed Deduplication Architecture for Ceph
Cachew ML input Data Processing as a Service
StRAID Stripe-threaded Architecture for Parity-based RAIDs with Ultra-fast SSDs
Pacman An Efficient Compaction Approach for Log-Structured Key-Value Store on Persistent Memory
Direct Access, High-Performance Memory Disaggregation with DirectCXL
NVMe SSD Failures in the Field the Fail-Stop and the Fail-Slow
FaaSNet Scalable and Fast Provisioning of Custom Serverless Container Runtimes at Alibaba Cloud Function Compute
WineFS a hugepage-aware file system for persistent memory that ages gracefully
KVIMR Key-Value Store Aware Data Management Middleware for Interlaced Magnetic Recording Based Hard Disk Drive
Scaling Large Production Clusters with Partitioned Synchronization
Characterizing and Optimizing Remote Persistent Memory with RDMA and NVM
Exploring the Design Space of Page Management for Multi-Tiered Memory Systems
NanoLog A Nanosecond Scale Logging System
OSCA An Online-Model Based Cache Allocation Scheme in Cloud Block Storage Systems
Differentiated Key-Value Storage Management for Balanced I/O Performance
Boosting Full-Node Repair in Erasure-Coded Storage
Jump-Starting Multivariate Time Series Anomaly Detection for Online Service Systems
ParaFS A Log-Structured File System to Exploit
HetPipe Enabling Large DNN Training on (Whimpy) Heterogeneous GPU Clusters through Integration of Pipelined Model Parallelism and Data Parallelism
Austere Flash Caching with Deduplication and Compression
BASTION A Security Enforcement Network Stack for Container Networks
MatrixKV Reducing Write Stalls and Write Amplification in LSM-tree Based KV Stores with a Matrix Container in NVM
A Deep Dive into DNS Query Failures
Parrot Efficient Serving of LLM-based Applications with Semantic Variable
MorphStore Analytical Query Engine with a Holistic Compression-Enabled Processing Model
OSCA An Online-Model Based Cache Allocation Scheme in Cloud Block Storage Systems
RackSched A Microsecond-Scale Scheduler for Rack-Scale Computers
PIDS Attribute Decomposition for Improved Compression and Query Performance in Columnar Storage
Cerebro A Data System for Optimized Deep Learning Model Selection
Microsecond Consensus for Microsecond Applications
How to Copy Files
FIRM An Intelligent Fine-grained Resource Management Framework for SLO-Oriented Microservices
PACEMAKER Avoiding HeART attacks in storage clusters with disk-adaptive redundancy
LinnOS Predictability on Unpredictable Flash Storage with a Light Neural Network
Sundial Fault-tolerant Clock Synchronization for Datacenters
From WiscKey to Bourbon- A Learned Index for Log-Structured Merge Trees
HetPipe Enabling Large DNN Training on (Whimpy) Heterogeneous GPU Clusters through Integration of Pipelined Model Parallelism and Data Parallelism
Fast RDMA-based Ordered Key-Value Store using Remote Learned Cache
A Unified Architecture for Accelerating Distributed DNN Training in Heterogeneous GPU/CPU Clusters
Austere Flash Caching with Deduplication and Compression
An Empirical Guide to the Behavior and Use of Scalable Persistent Memory
BASTION A Security Enforcement Network Stack for Container Networks
GIFT A Coupon Based Throttle-and-Reward Mechanism for Fair and Efficient I/O Bandwidth Management on Parallel Storage Systems
MatrixKV Reducing Write Stalls and Write Amplification in LSM-tree Based KV Stores with a Matrix Container in NVM
A Deep Dive into DNS Query Failures
OmniCache Collaborative Caching for Near-storage Accelerators
In-Memory Key-Value Store Live Migration with NetMigrate
MinFlow High-performance and Cost-efficient Data Passing for I/O-intensive Stateful Serverless Analytics
We Ain’t Afraid of No File Fragmentation Causes and Prevention of Its Performance Impact on Modern Flash SSDs
MiDAS Minimizing Write Amplification in Log-Structured Systems through Adaptive Group Number and Size Configuration
Symbiosis The Art of Application and Kernel Cache Cooperation
What’s the Story in EBS Glory Evolutions and Lessons in Building Cloud Block Store
FAST 24 论文速览
ELECT Enabling Erasure Coding Tiering for LSM-tree-based Storage
Fisc A Large-scale Cloud-native-oriented File System
Multi-view Feature-based SSD Failure Prediction What, When, and Why
SMRStore A Storage Engine for Cloud Object Storage on HM-SMR Drive
HadaFS A File System Bridging the Local and Shared Burst Buffer for Exascale Supercomputers
InftyDedup Scalable and Cost-Effective Cloud Tiering with Deduplication
Integrated Host-SSD Mapping Table Management for Improving User Experience of Smartphones
Perseus A Fail-Slow Detection Framework for Cloud Storage Systems
GL-Cache Group-level learning for efficient and high-performance caching
ADOC Automatically Harmonizing Dataflow Between Components in Log-Structured Key-Value Stores for Improved Performance
ParaRC Embracing Sub-Packetization for Repair Parallelization in MSR-Coded Storage
Hydra Resilient and Highly Available Remote Memory
ctFS Replacing File Indexing with Hardware Memory Translation through Contiguous File Allocation for Persistent Memory
Operational Characteristics of SSDs in Enterprise Storage Systems A Large-Scale Field Study
DEPART Replica Decoupling for Distributed Key-Value Storage
TVStore Automatically Bounding Time Series Storage via Time-Varying Compression
DeepSketch A New Machine Learning-Based Reference Search Technique for Post-Deduplication Delta Compression
Improving the Reliability of Next Generation SSDs using WOM-v Codes
InfiniFS An Efficient Metadata Service for Large-Scale Distributed Filesystems
The what, The from, and The to The Migration Games in Deduplicated Systems
Closing the B+ -tree vs LSM-tree Write Amplification Gap on Modern Storage Hardware with Built-in Transparent Compression
Tiny-Tail Flash Near-Perfect Elimination of Garbage Collection Tail Latencies in NAND SSDs
D2FQ Device-Direct Fair Queueing for NVMe SSDs
FlashNeuron SSD-Enabled Large-Batch Training of Very Deep Neural Networks
Optimizing Storage Performance with Calibrated Interrupts
Aurogon Taming Aborts in All Phases for Distributed In-Memory Transactions
Clay Codes Moulding MDS Codes to Yield an MSR Code
ROART Range-query Optimized Persistent ART
REMIX Efficient Range Query for LSM-trees
CheckFreq Frequent, Fine-Grained DNN Checkpointing
Rethinking File Mapping for Persistent Memory
The Storage Hierarchy is Not a Hierarchy Optimizing Caching on Modern Storage Devices with Orthus
Facebook’s Tectonic Filesystem Efficiency from Exascale
Exploiting Combined Locality for Wide-Stripe Erasure Coding in Distributed Storage
SpanDB A Fast, Cost-Effective LSM-tree Based KV Store on Hybrid Storage
How to Copy Files
DistCache Provable Load Balancing for LargeScale Storage Systems with Distributed Caching
OpenEC Toward Unified and Configurable Erasure Coding Management in Distributed Storage Systems
An Empirical Guide to the Behavior and Use of Scalable Persistent Memory
GIFT A Coupon Based Throttle-and-Reward Mechanism for Fair and Efficient I/O Bandwidth Management on Parallel Storage Systems
Burstable Cloud Block Storage with Data Processing Units
Optimizing Resource Allocation in Hyperscale Datacenters Scalability, Usability, and Experiences
μSlope High Compression and Fast Search on Semi-Structured Logs
Parrot Efficient Serving of LLM-based Applications with Semantic Variable
Sabre Hardware-Accelerated Snapshot Compression for Serverless MicroVMs
VBase Unifying Online Vector Similarity Search and Relational Queries via Relaxed Monotonicity
ServerlessLLM Locality-Enhanced Serverless Inference for Large Language Models
RON One-Way Circular Shortest Routing to Achieve Efficient and Bounded-waiting Spinlocks
Take Out the TraChe Maximizing (Tra)nsactional Ca(che) Hit Rate
eZNS An Elastic Zoned Namespace for Commodity ZNS SSDs
Ensō A Streaming Interface for NIC-Application Communication
Orca A Distributed Serving System for Transformer-Based Generative Models
Hyrax Fail-in-Place Server Operation in Cloud Platforms
Hubble Performance Debugging with In-Production, Just-In-Time Method Tracing on Android
Looking Beyond GPUs for DNN Scheduling on Multi-Tenant Clusters
Demystifying and Checking Silent Semantic Violations in Large Distributed Systems
Carbink Fault-Tolerant Far Memory
XRP In-Kernel Storage Functions with eBPF
Owl Scale and Flexibility in Distribution of Hot Content
Tiger Disk-Adaptive Redundancy Without Placement Restrictions
Metastable Failures in the Wild
BlockFlex Enabling Storage Harvesting with Software-Defined Flash in Modern Cloud Platforms
TriCache A User-Transparent Block Enabling High-Performance Out-of-Core Processing with In-Memory Programs
Microsecond-scale Preemption for Concurrent GPU-accelerated DNN Inferences
Modernizing File System through In-Storage Indexing
ZNS+ Advanced Zoned Namespace Interface for Supporting In-Storage Zone Compaction
Privacy Budget Scheduling
CLP Efficient and Scalable Search on Compressed Text Logs
Oort Efficient Federated Learning via Guided Participant Selection
Rearchitecting Linux Storage Stack for µs Latency and High Throughput
RackSched A Microsecond-Scale Scheduler for Rack-Scale Computers
Gandiva Introspective Cluster Scheduling for Deep Learning
Microsecond Consensus for Microsecond Applications
FIRM An Intelligent Fine-grained Resource Management Framework for SLO-Oriented Microservices
PACEMAKER Avoiding HeART attacks in storage clusters with disk-adaptive redundancy
LinnOS Predictability on Unpredictable Flash Storage with a Light Neural Network
Sundial Fault-tolerant Clock Synchronization for Datacenters
From WiscKey to Bourbon- A Learned Index for Log-Structured Merge Trees
Fast RDMA-based Ordered Key-Value Store using Remote Learned Cache
A Unified Architecture for Accelerating Distributed DNN Training in Heterogeneous GPU/CPU Clusters
DistCache Provable Load Balancing for LargeScale Storage Systems with Distributed Caching
OpenEC Toward Unified and Configurable Erasure Coding Management in Distributed Storage Systems
An Empirical Evaluation of Columnar Storage Formats
Elf Erasing-based Lossless Floating-Point Compression
Chimp Efficient Lossless Floating Point Compression for Time Series Databases
Time Series Data Encoding for Efficient Storage A Comparative Analysis in Apache IoTDB
TencentCLS The Cloud Log Service with High Query Performances
MorphStore Analytical Query Engine with a Holistic Compression-Enabled Processing Model
Tsunami A Learned Multi-dimensional Index for Correlated Data and Skewed Workloads
PIDS Attribute Decomposition for Improved Compression and Query Performance in Columnar Storage
Cerebro A Data System for Optimized Deep Learning Model Selection
Analyzing and Mitigating Data Stalls in DNN Training
Shard Manager A Generic Shard Management Framework for Geo-distributed Applications
RAIZN Redundant Array of Independent Zoned Namespaces
LogGrep
Zeus Locality-aware Distributed Transactions
Shard Manager A Generic Shard Management Framework for Geo-distributed Applications
Geometric Partitioning Explore the Boundary of Optimal Erasure Code Repair
Good to the Last Bit Data-Driven Encoding with CodecDB
Where did my 256 GB go? A Measurement Analysis of Storage Consumption on Smart Mobile Devices
Adaptive Compression for Fast scans on string columns
DeltaFS A Scalable No-Ground-Truth Filesystem For Massively-Parallel Computing
JSON Tiles Fast Analytics on Semi-Structured Data
Accelerating XOR-based erasure coding using program optimization techniques
iVPF Numerical Invertible Volume Preserving Flow for Efficient Lossless Compression
FaaSNet Scalable and Fast Provisioning of Custom Serverless Container Runtimes at Alibaba Cloud Function Compute
LogECMem Coupling Erasure-Coded In-Memory Key-Value Stores with Parity Logging
Modernizing File System through In-Storage Indexing
Witcher Systematic Crash Consistency Testing for Non-Volatile Memory Key-Value Stores
Tsunami A Learned Multi-dimensional Index for Correlated Data and Skewed Workloads
VF2Boost Very Fast Vertical Federated Gradient Boosting for Cross-Enterprise Learning
WineFS a hugepage-aware file system for persistent memory that ages gracefully
KVIMR Key-Value Store Aware Data Management Middleware for Interlaced Magnetic Recording Based Hard Disk Drive
IODA A Host/Device Co-Design for Strong Predictability Contract on Modern Flash Storage
D2FQ Device-Direct Fair Queueing for NVMe SSDs
Scaling Large Production Clusters with Partitioned Synchronization
FragPicker A New Defragmentation Tool for Modern Storage Devices
ZNS+ Advanced Zoned Namespace Interface for Supporting In-Storage Zone Compaction
Characterizing and Optimizing Remote Persistent Memory with RDMA and NVM
Kangaroo Caching Billions of Tiny Objects on Flash
FlashNeuron SSD-Enabled Large-Batch Training of Very Deep Neural Networks
Exploring the Design Space of Page Management for Multi-Tiered Memory Systems
Optimizing Storage Performance with Calibrated Interrupts
Differentiated Key-Value Storage Management for Balanced I/O Performance
Boosting Full-Node Repair in Erasure-Coded Storage
Privacy Budget Scheduling
Jump-Starting Multivariate Time Series Anomaly Detection for Online Service Systems
Bridging the Performance Gap for Copy-based Garbage Collectors atop Non-Volatile Memory
CLP Efficient and Scalable Search on Compressed Text Logs
Oort Efficient Federated Learning via Guided Participant Selection
Rearchitecting Linux Storage Stack for µs Latency and High Throughput
Achieving Low Tail-latency and High Scalability for Serializable Transactions in Edge Computing
ROART Range-query Optimized Persistent ART
REMIX Efficient Range Query for LSM-trees
CheckFreq Frequent, Fine-Grained DNN Checkpointing
Rethinking File Mapping for Persistent Memory
Near-Optimal Latency Versus Cost Tradeoffs in Geo-Distributed Storage
The Storage Hierarchy is Not a Hierarchy Optimizing Caching on Modern Storage Devices with Orthus
Facebook’s Tectonic Filesystem Efficiency from Exascale
Exploiting Combined Locality for Wide-Stripe Erasure Coding in Distributed Storage
SpanDB A Fast, Cost-Effective LSM-tree Based KV Store on Hybrid Storage
Analyzing and Mitigating Data Stalls in DNN Training
Concordia Distributed Shared Memory with In-Network Cache Coherence
Known Knowns and Unknowns Near-realtime Earth Observation Via Query Bifurcation in Serval
Seer Enabling Future-Aware Online Caching in Networked Systems
Characterization of Large Language Model Development in the Datacenter
Unlocking unallocated cloud capacity for long, uninterruptible workloads
RDMA is Turing complete, we just did not know it yet!
Near-Optimal Latency Versus Cost Tradeoffs in Geo-Distributed Storage
Albis High-Performance File Format for Big Data Systems
NanoLog A Nanosecond Scale Logging System
Clay Codes Moulding MDS Codes to Yield an MSR Code
Gandiva Introspective Cluster Scheduling for Deep Learning
CFS Scaling Metadata Service for Distributed File System via Pruned Scope of Critical Sections
Eurosys 23 论文速递
Hi-Speed DNN Training with Espresso Unleashing the Full Potential of Gradient Compression with Near-Optimal Usage Strategies
ResPCT Fast Checkpointing in Non-volatile Memory for Multi-threaded Applications
LogGrep
Zeus Locality-aware Distributed Transactions
Achieving Low Tail-latency and High Scalability for Serializable Transactions in Edge Computing
降低宽列纠删码的长尾延迟
轻量级神经网络超参数调优并行训练框架
宽列纠删码存储的设计与实现
基于超算的RSA分解因子平台搭建
ParaFS A Log-Structured File System to Exploit
TRINITY A Fast Compressed Multi-attribute Data Store
p2KVS a Portable 2-Dimensional Parallelizing Framework to Improve Scalability of Key-value Stores on SSDs
Bridging the Performance Gap for Copy-based Garbage Collectors atop Non-Volatile Memory
Chimp Efficient Lossless Floating Point Compression for Time Series Databases
Orca A Distributed Serving System for Transformer-Based Generative Models
Time Series Data Encoding for Efficient Storage A Comparative Analysis in Apache IoTDB
Cachew ML input Data Processing as a Service
TencentCLS The Cloud Log Service with High Query Performances
StRAID Stripe-threaded Architecture for Parity-based RAIDs with Ultra-fast SSDs
Hubble Performance Debugging with In-Production, Just-In-Time Method Tracing on Android
Looking Beyond GPUs for DNN Scheduling on Multi-Tenant Clusters
Demystifying and Checking Silent Semantic Violations in Large Distributed Systems
ResPCT Fast Checkpointing in Non-volatile Memory for Multi-threaded Applications
Carbink Fault-Tolerant Far Memory
XRP In-Kernel Storage Functions with eBPF
Owl Scale and Flexibility in Distribution of Hot Content
Tiger Disk-Adaptive Redundancy Without Placement Restrictions
CompressDB Enabling Efficient Compressed Data Direct Processing for Various Databases
Hydra Resilient and Highly Available Remote Memory
Metastable Failures in the Wild
Pacman An Efficient Compaction Approach for Log-Structured Key-Value Store on Persistent Memory
Direct Access, High-Performance Memory Disaggregation with DirectCXL
BlockFlex Enabling Storage Harvesting with Software-Defined Flash in Modern Cloud Platforms
TriCache A User-Transparent Block Enabling High-Performance Out-of-Core Processing with In-Memory Programs
Microsecond-scale Preemption for Concurrent GPU-accelerated DNN Inferences
NVMe SSD Failures in the Field the Fail-Stop and the Fail-Slow
ctFS Replacing File Indexing with Hardware Memory Translation through Contiguous File Allocation for Persistent Memory
p2KVS a Portable 2-Dimensional Parallelizing Framework to Improve Scalability of Key-value Stores on SSDs
Operational Characteristics of SSDs in Enterprise Storage Systems A Large-Scale Field Study
DEPART Replica Decoupling for Distributed Key-Value Storage
TVStore Automatically Bounding Time Series Storage via Time-Varying Compression
DeepSketch A New Machine Learning-Based Reference Search Technique for Post-Deduplication Delta Compression
Improving the Reliability of Next Generation SSDs using WOM-v Codes
InfiniFS An Efficient Metadata Service for Large-Scale Distributed Filesystems
The what, The from, and The to The Migration Games in Deduplicated Systems
RDMA is Turing complete, we just did not know it yet!
Closing the B+ -tree vs LSM-tree Write Amplification Gap on Modern Storage Hardware with Built-in Transparent Compression
Aurogon Taming Aborts in All Phases for Distributed In-Memory Transactions
FIFO Queues are all You Need for Cache Eviction
Gemini Fast Failure Recovery in Distributed Training with In-Memory Checkpoints
Enabling High-Performance and Secure Userspace NVM File Systems with the TRIO Architecture
A Cloud-Scale Characterization of Remote Procedure Calls
XFaaS Hyperscale and Low Cost Serverless Functions at Meta
TreeSLS A Whole-system Persistent Microkernel with Tree-structured State Checkpoint on NVM
SOSP 23 论文速览
Efficient Memory Management for Large Language Model Serving with PagedAttention
Shard Manager A Generic Shard Management Framework for Geo-distributed Applications
Shard Manager A Generic Shard Management Framework for Geo-distributed Applications
Geometric Partitioning Explore the Boundary of Optimal Erasure Code Repair
Witcher Systematic Crash Consistency Testing for Non-Volatile Memory Key-Value Stores
IODA A Host/Device Co-Design for Strong Predictability Contract on Modern Flash Storage
FragPicker A New Defragmentation Tool for Modern Storage Devices
Kangaroo Caching Billions of Tiny Objects on Flash
LogGrep+ improve retrieval efficiency on highly compressed cloud log with encoding aware schemes
Neural Compression Review
宽列纠删码
存储盘用户画像刻画
What is DNA Storage
Tiny-Tail Flash Near-Perfect Elimination of Garbage Collection Tail Latencies in NAND SSDs
VF2Boost Very Fast Vertical Federated Gradient Boosting for Cross-Enterprise Learning
Design Considerations and Analysis of Multi-Level Erasure Coding in Large-Scale Data Centers
DeltaFS A Scalable No-Ground-Truth Filesystem For Massively-Parallel Computing
Accelerating XOR-based erasure coding using program optimization techniques
LogECMem Coupling Erasure-Coded In-Memory Key-Value Stores with Parity Logging
iVPF Numerical Invertible Volume Preserving Flow for Efficient Lossless Compression
BTRBLOCKS Efficient Columnar Compression for Data Lakes
Selection Pushdown in Colum Stores using Bit Manipulation Instructions
CompressDB Enabling Efficient Compressed Data Direct Processing for Various Databases
Good to the Last Bit Data-Driven Encoding with CodecDB
Adaptive Compression for Fast scans on string columns
JSON Tiles Fast Analytics on Semi-Structured Data
Where did my 256 GB go? A Measurement Analysis of Storage Consumption on Smart Mobile Devices
FIFO Queues are all You Need for Cache Eviction
VBase Unifying Online Vector Similarity Search and Relational Queries via Relaxed Monotonicity
Gemini Fast Failure Recovery in Distributed Training with In-Memory Checkpoints
BTRBLOCKS Efficient Columnar Compression for Data Lakes
Enabling High-Performance and Secure Userspace NVM File Systems with the TRIO Architecture
RON One-Way Circular Shortest Routing to Achieve Efficient and Bounded-waiting Spinlocks
Nodens Enabling Resource Efficient and Fast QoS Recovery of Dynamic Microservice Applications in Datacenters
A Cloud-Scale Characterization of Remote Procedure Calls
Calcspar A Contract-Aware LSM Store for Cloud Storage with Low Latency Spikes
Elf Erasing-based Lossless Floating-Point Compression
XFaaS Hyperscale and Low Cost Serverless Functions at Meta
Take Out the TraChe Maximizing (Tra)nsactional Ca(che) Hit Rate
Design Considerations and Analysis of Multi-Level Erasure Coding in Large-Scale Data Centers
Language Model is Compression
eZNS An Elastic Zoned Namespace for Commodity ZNS SSDs
TreeSLS A Whole-system Persistent Microkernel with Tree-structured State Checkpoint on NVM
SOSP 23 论文速览
Ensō A Streaming Interface for NIC-Application Communication
Efficient Memory Management for Large Language Model Serving with PagedAttention
Light-Dedup A Light-weight Inline Deduplication Framework for Non-Volatile Memory File Systems
LLFree Scalable and Optionally-Persistent Page-Frame Allocation
Explore Data Placement Algorithm for Balanced Recovery Load Distribution
TiDedup A New Distributed Deduplication Architecture for Ceph
Selection Pushdown in Colum Stores using Bit Manipulation Instructions
Fisc A Large-scale Cloud-native-oriented File System
Hyrax Fail-in-Place Server Operation in Cloud Platforms
Multi-view Feature-based SSD Failure Prediction What, When, and Why
Unlocking unallocated cloud capacity for long, uninterruptible workloads
SMRStore A Storage Engine for Cloud Object Storage on HM-SMR Drive
HadaFS A File System Bridging the Local and Shared Burst Buffer for Exascale Supercomputers
CFS Scaling Metadata Service for Distributed File System via Pruned Scope of Critical Sections
Eurosys 23 论文速递
InftyDedup Scalable and Cost-Effective Cloud Tiering with Deduplication
Integrated Host-SSD Mapping Table Management for Improving User Experience of Smartphones
Perseus A Fail-Slow Detection Framework for Cloud Storage Systems
GL-Cache Group-level learning for efficient and high-performance caching
ADOC Automatically Harmonizing Dataflow Between Components in Log-Structured Key-Value Stores for Improved Performance
ParaRC Embracing Sub-Packetization for Repair Parallelization in MSR-Coded Storage
Hi-Speed DNN Training with Espresso Unleashing the Full Potential of Gradient Compression with Near-Optimal Usage Strategies
A Quantitative Analysis and Guidelines of Data Streaming Accelerator in Modern Intel Xeon Scalable Processors
RAIZN Redundant Array of Independent Zoned Namespaces
Zebra ZRWA-EnaBled Redundant Array of Zoned Namespace SSDs
面向新型计算应用构建存储系统的若干思考
SDC Study
MedCompressor：医学图像压缩
Language Model is Compression
Known Knowns and Unknowns Near-realtime Earth Observation Via Query Bifurcation in Serval
Burstable Cloud Block Storage with Data Processing Units
OmniCache Collaborative Caching for Near-storage Accelerators
Ethane An Asymmetric File System for Disaggregated Persistent Memory
Seer Enabling Future-Aware Online Caching in Networked Systems
Scalable Billion-point Approximate Nearest Neighbor Search Using SmartSSDs
Optimizing Resource Allocation in Hyperscale Datacenters Scalability, Usability, and Experiences
μSlope High Compression and Fast Search on Semi-Structured Logs
Sabre Hardware-Accelerated Snapshot Compression for Serverless MicroVMs
Removing Obstacles before Breaking Through the Memory Wall A Close Look at HBM Errors in the Field
An Empirical Study of Rust-for-Linux The Success, Dissatisfaction, and Compromise
ATC '24 经历分享
Characterization of Large Language Model Development in the Datacenter
In-Memory Key-Value Store Live Migration with NetMigrate
StreamCache Revisiting Page Cache for File Scanning on Fast Storage Devices
MinFlow High-performance and Cost-efficient Data Passing for I/O-intensive Stateful Serverless Analytics
ServerlessLLM Locality-Enhanced Serverless Inference for Large Language Models
TRINITY A Fast Compressed Multi-attribute Data Store
A Quantitative Analysis and Guidelines of Data Streaming Accelerator in Modern Intel Xeon Scalable Processors
We Ain’t Afraid of No File Fragmentation Causes and Prevention of Its Performance Impact on Modern Flash SSDs
MiDAS Minimizing Write Amplification in Log-Structured Systems through Adaptive Group Number and Size Configuration
Symbiosis The Art of Application and Kernel Cache Cooperation
What’s the Story in EBS Glory Evolutions and Lessons in Building Cloud Block Store
FAST 24 论文速览
ELECT Enabling Erasure Coding Tiering for LSM-tree-based Storage
An Empirical Evaluation of Columnar Storage Formats