DNS
Network
Bottleneck
storage structure
container network
persistent memory
NVM
NVSL lab
SSD
Erasure Code
Pipeline NN
data structure
transaction system
time synchronize
micro service
machine learning
cache
file system
RDMA
replication system
DNN
optimization
Deep Learning
model selection
erasure code
wide stripe
storage system
latency
PM
checkpoint
GPU
LSM tree；
Data Structure
Operating system
Rack
Schedule
edge calculation
logging
linux kernel
block layer
scheduler
Federated Learning
log
log search
GC
Java
ML
privacy
LSM tree
Transaction
System
Cache
OS
fragment
DNA Storage
IMR Disk
Liberation
AI
Crash Consistency
File System
KV
Turing Complete
compression
Compression
EC
Compile
Study
FS
metadata
Storage
storage
study paper
database
Persistent Memory
NVMe
Fault Tolerance
FTL
Open channel SSD
CXL
transaction
Wide Stripe EC
DB
Erasure Coding
eBPF
Checkpoint
study
JIT
Tracing
Debugging
RAID
search engine
Machine Learning
Clay Code
KV Store
Fail-Slow
ZNS
SDC
CPU
Smart Phone
Deduplication
Metadata
HPC
Burst Buffer
SMR
Serverless
Virtual Machine
Reliability
Failure Prediction
Data Management
Kernel
Page Allocation
ZNS SSD
LLM
microkernel
serverless
LSM
tail latency
RPC
Microservice
Micro-architecture
Security
Database
KV store
Checkpointing
Block Store
Deployed System
Log-structured System
- DNS 1
- Network 3
- Bottleneck 1
- storage structure 2
- 153
- container network 1
- persistent memory 1
- NVM 10
- NVSL lab 1
- SSD 8
- Erasure Code 9
- Pipeline NN 1
- data structure 2
- transaction system 1
- time synchronize 1
- micro service 1
- machine learning 1
- cache 3
- file system 2
- RDMA 3
- replication system 1
- DNN 3
- optimization 1
- Deep Learning 1
- model selection 1
- erasure code 2
- wide stripe 1
- storage system 1
- latency 1
- PM 5
- checkpoint 1
- GPU 2
- LSM tree； 1
- Data Structure 1
- Operating system 1
- Rack 1
- Schedule 1
- edge calculation 1
- logging 1
- linux kernel 1
- block layer 1
- scheduler 1
- Federated Learning 1
- log 4
- log search 1
- GC 2
- Java 1
- ML 8
- privacy 1
- LSM tree 3
- Transaction 1
- System 2
- Cache 6
- OS 1
- fragment 1
- DNA Storage 1
- IMR Disk 1
- Liberation 1
- AI 5
- Crash Consistency 1
- File System 5
- KV 3
- Turing Complete 1
- compression 6
- Compression 10
- EC 1
- Compile 1
- Study 3
- FS 1
- metadata 1
- Storage 1
- storage 1
- study paper 2
- database 1
- Persistent Memory 1
- NVMe 1
- Fault Tolerance 5
- FTL 1
- Open channel SSD 1
- CXL 1
- transaction 1
- Wide Stripe EC 1
- DB 1
- Erasure Coding 2
- eBPF 1
- Checkpoint 1
- study 1
- JIT 1
- Tracing 1
- Debugging 1
- RAID 4
- search engine 1
- Machine Learning 3
- Clay Code 1
- KV Store 1
- Fail-Slow 1
- ZNS 1
- SDC 1
- CPU 1
- Smart Phone 1
- Deduplication 3
- Metadata 1
- HPC 1
- Burst Buffer 1
- SMR 1
- Serverless 1
- Virtual Machine 1
- Reliability 2
- Failure Prediction 1
- Data Management 1
- Kernel 1
- Page Allocation 1
- ZNS SSD 2
- LLM 2
- microkernel 1
- serverless 1
- LSM 1
- tail latency 1
- RPC 1
- Microservice 1
- Micro-architecture 1
- Security 1
- Database 1
- KV store 1
- Checkpointing 1
- Block Store 1
- Deployed System 1
- Log-structured System 1
DNS
Network
- » A Cloud-Scale Characterization of Remote Procedure Calls SOSP23
- » Ensō A Streaming Interface for NIC-Application Communication OSDI23
- » A Deep Dive into DNS Query Failures ATC20
Bottleneck
storage structure
- » From WiscKey to Bourbon- A Learned Index for Log-Structured Merge Trees OSDI20
- » MatrixKV Reducing Write Stalls and Write Amplification in LSM-tree Based KV Stores with a Matrix Container in NVM ATC20
- » FAST 24 论文速览 FAST24
- » ELECT Enabling Erasure Coding Tiering for LSM-tree-based Storage FAST24
- » An Empirical Evaluation of Columnar Storage Formats VLDB24
- » BTRBLOCKS Efficient Columnar Compression for Data Lakes SIGMOD23
- » Enabling High-Performance and Secure Userspace NVM File Systems with the TRIO Architecture SOSP23
- » RON One-Way Circular Shortest Routing to Achieve Efficient and Bounded-waiting Spinlocks OSDI23
- » Nodens Enabling Resource Efficient and Fast QoS Recovery of Dynamic Microservice Applications in Datacenters ATC23
- » A Cloud-Scale Characterization of Remote Procedure Calls SOSP23
- » Calcspar A Contract-Aware LSM Store for Cloud Storage with Low Latency Spikes ATC23
- » XFaaS Hyperscale and Low Cost Serverless Functions at Meta SOSP23
- » Take Out the TraChe Maximizing (Tra)nsactional Ca(che) Hit Rate OSDI23
- » Design Considerations and Analysis of Multi-Level Erasure Coding in Large-Scale Data Centers SC23
- » Language Model is Compression Arxiv23
- » eZNS An Elastic Zoned Namespace for Commodity ZNS SSDs OSDI23
- » TreeSLS A Whole-system Persistent Microkernel with Tree-structured State Checkpoint on NVM SOSP23
- » SOSP 23 论文速览 SOSP23
- » Efficient Memory Management for Large Language Model Serving with PagedAttention SOSP23
- » Light-Dedup A Light-weight Inline Deduplication Framework for Non-Volatile Memory File Systems ATC23
- » Zebra ZRWA-EnaBled Redundant Array of Zoned Namespace SSDs 科研分享
- » LLFree Scalable and Optionally-Persistent Page-Frame Allocation ATC23
- » Explore Data Placement Algorithm for Balanced Recovery Load Distribution ATC23
- » 面向新型计算应用构建存储系统的若干思考 科研分享
- » TiDedup A New Distributed Deduplication Architecture for Ceph ATC23
- » MedCompressor：医学图像压缩 科研分析
- » Chimp Efficient Lossless Floating Point Compression for Time Series Databases VLDB22
- » Selection Pushdown in Colum Stores using Bit Manipulation Instructions SIGMOD23
- » Fisc A Large-scale Cloud-native-oriented File System FAST23
- » Orca A Distributed Serving System for Transformer-Based Generative Models OSDI22
- » Hyrax Fail-in-Place Server Operation in Cloud Platforms OSDI23
- » Multi-view Feature-based SSD Failure Prediction What, When, and Why FAST23
- » Unlocking unallocated cloud capacity for long, uninterruptible workloads NSDI23
- » SMRStore A Storage Engine for Cloud Object Storage on HM-SMR Drive FAST23
- » HadaFS A File System Bridging the Local and Shared Burst Buffer for Exascale Supercomputers FAST23
- » Shard Manager A Generic Shard Management Framework for Geo-distributed Applications SOSP21
- » CFS Scaling Metadata Service for Distributed File System via Pruned Scope of Critical Sections Eurosys23
- » Time Series Data Encoding for Efficient Storage A Comparative Analysis in Apache IoTDB VLDB22
- » Eurosys 23 论文速递 Eurosys23
- » InftyDedup Scalable and Cost-Effective Cloud Tiering with Deduplication FAST23
- » Integrated Host-SSD Mapping Table Management for Improving User Experience of Smartphones FAST23
- » Cachew ML input Data Processing as a Service ATC22
- » SDC Study 科研分享
- » RAIZN Redundant Array of Independent Zoned Namespaces ASPLOS21
- » Perseus A Fail-Slow Detection Framework for Cloud Storage Systems FAST23
- » GL-Cache Group-level learning for efficient and high-performance caching FAST23
- » ADOC Automatically Harmonizing Dataflow Between Components in Log-Structured Key-Value Stores for Improved Performance FAST23
- » ParaRC Embracing Sub-Packetization for Repair Parallelization in MSR-Coded Storage FAST23
- » Hi-Speed DNN Training with Espresso Unleashing the Full Potential of Gradient Compression with Near-Optimal Usage Strategies Eurosys23
- » TencentCLS The Cloud Log Service with High Query Performances VLDB22
- » StRAID Stripe-threaded Architecture for Parity-based RAIDs with Ultra-fast SSDs ATC22
- » Hubble Performance Debugging with In-Production, Just-In-Time Method Tracing on Android OSDI22
- » Looking Beyond GPUs for DNN Scheduling on Multi-Tenant Clusters OSDI22
- » Demystifying and Checking Silent Semantic Violations in Large Distributed Systems OSDI22
- » LogGrep+ improve retrieval efficiency on highly compressed cloud log with encoding aware schemes nan
- » ResPCT Fast Checkpointing in Non-volatile Memory for Multi-threaded Applications Eurosys22
- » Carbink Fault-Tolerant Far Memory OSDI22
- » XRP In-Kernel Storage Functions with eBPF OSDI22
- » Owl Scale and Flexibility in Distribution of Hot Content OSDI22
- » Tiger Disk-Adaptive Redundancy Without Placement Restrictions OSDI22
- » Neural Compression Review nan
- » CompressDB Enabling Efficient Compressed Data Direct Processing for Various Databases SIGMOD22
- » Hydra Resilient and Highly Available Remote Memory FAST22
- » LogGrep Eurosys21
- » 宽列纠删码 nan
- » Zeus Locality-aware Distributed Transactions Eurosys21
- » Metastable Failures in the Wild OSDI22
- » Pacman An Efficient Compaction Approach for Log-Structured Key-Value Store on Persistent Memory ATC22
- » Direct Access, High-Performance Memory Disaggregation with DirectCXL ATC22
- » BlockFlex Enabling Storage Harvesting with Software-Defined Flash in Modern Cloud Platforms OSDI22
- » TriCache A User-Transparent Block Enabling High-Performance Out-of-Core Processing with In-Memory Programs OSDI22
- » Shard Manager A Generic Shard Management Framework for Geo-distributed Applications SOSP21
- » Geometric Partitioning Explore the Boundary of Optimal Erasure Code Repair SOSP21
- » Microsecond-scale Preemption for Concurrent GPU-accelerated DNN Inferences OSDI22
- » MorphStore Analytical Query Engine with a Holistic Compression-Enabled Processing Model VLDB20
- » NVMe SSD Failures in the Field the Fail-Stop and the Fail-Slow ATC22
- » ctFS Replacing File Indexing with Hardware Memory Translation through Contiguous File Allocation for Persistent Memory FAST22
- » Good to the Last Bit Data-Driven Encoding with CodecDB SIGMOD21
- » Where did my 256 GB go? A Measurement Analysis of Storage Consumption on Smart Mobile Devices SIGMETRIC21
- » Adaptive Compression for Fast scans on string columns SIGMOD21
- » p2KVS a Portable 2-Dimensional Parallelizing Framework to Improve Scalability of Key-value Stores on SSDs EuroSys22
- » 存储盘用户画像刻画 nan
- » Operational Characteristics of SSDs in Enterprise Storage Systems A Large-Scale Field Study FAST22
- » DEPART Replica Decoupling for Distributed Key-Value Storage FAST22
- » TVStore Automatically Bounding Time Series Storage via Time-Varying Compression FAST22
- » JSON Tiles Fast Analytics on Semi-Structured Data SIGMOD21
- » Accelerating XOR-based erasure coding using program optimization techniques SC21
- » DeepSketch A New Machine Learning-Based Reference Search Technique for Post-Deduplication Delta Compression FAST22
- » Improving the Reliability of Next Generation SSDs using WOM-v Codes FAST22
- » InfiniFS An Efficient Metadata Service for Large-Scale Distributed Filesystems FAST22
- » iVPF Numerical Invertible Volume Preserving Flow for Efficient Lossless Compression CVPR21
- » The what, The from, and The to The Migration Games in Deduplicated Systems FAST22
- » RDMA is Turing complete, we just did not know it yet! NSDI22
- » Closing the B+ -tree vs LSM-tree Write Amplification Gap on Modern Storage Hardware with Built-in Transparent Compression FAST22
- » FaaSNet Scalable and Fast Provisioning of Custom Serverless Container Runtimes at Alibaba Cloud Function Compute ATC21
- » LogECMem Coupling Erasure-Coded In-Memory Key-Value Stores with Parity Logging SC21
- » Modernizing File System through In-Storage Indexing OSDI21
- » Witcher Systematic Crash Consistency Testing for Non-Volatile Memory Key-Value Stores SOSP21
- » KVIMR Key-Value Store Aware Data Management Middleware for Interlaced Magnetic Recording Based Hard Disk Drive ATC21
- » What is DNA Storage nan
- » IODA A Host/Device Co-Design for Strong Predictability Contract on Modern Flash Storage SOSP21
- » D2FQ Device-Direct Fair Queueing for NVMe SSDs FAST21
- » Scaling Large Production Clusters with Partitioned Synchronization ATC21
- » ZNS+ Advanced Zoned Namespace Interface for Supporting In-Storage Zone Compaction OSDI21
- » Characterizing and Optimizing Remote Persistent Memory with RDMA and NVM ATC21
- » Kangaroo Caching Billions of Tiny Objects on Flash SOSP21
- » FlashNeuron SSD-Enabled Large-Batch Training of Very Deep Neural Networks FAST21
- » Exploring the Design Space of Page Management for Multi-Tiered Memory Systems ATC21
- » NanoLog A Nanosecond Scale Logging System ATC18
- » OSCA An Online-Model Based Cache Allocation Scheme in Cloud Block Storage Systems ATC20
- » Optimizing Storage Performance with Calibrated Interrupts FAST21
- » Aurogon Taming Aborts in All Phases for Distributed In-Memory Transactions FAST22
- » Differentiated Key-Value Storage Management for Balanced I/O Performance ATC21
- » Boosting Full-Node Repair in Erasure-Coded Storage ATC21
- » Privacy Budget Scheduling OSDI21
- » Jump-Starting Multivariate Time Series Anomaly Detection for Online Service Systems ATC21
- » Bridging the Performance Gap for Copy-based Garbage Collectors atop Non-Volatile Memory EuroSys21
- » CLP Efficient and Scalable Search on Compressed Text Logs OSDI21
- » Oort Efficient Federated Learning via Guided Participant Selection OSDI21
- » Rearchitecting Linux Storage Stack for µs Latency and High Throughput OSDI21
- » Clay Codes Moulding MDS Codes to Yield an MSR Code FAST18
- » ParaFS A Log-Structured File System to Exploit ATC16
- » 降低宽列纠删码的长尾延迟 毕业设计
- » 轻量级神经网络超参数调优并行训练框架 毕业设计
- » 宽列纠删码存储的设计与实现 毕业设计
- » 基于超算的RSA分解因子平台搭建 毕业设计
- » Achieving Low Tail-latency and High Scalability for Serializable Transactions in Edge Computing Eurosys21
- » RackSched A Microsecond-Scale Scheduler for Rack-Scale Computers OSDI20
- » ROART Range-query Optimized Persistent ART FAST21
- » Gandiva Introspective Cluster Scheduling for Deep Learning OSDI18
- » CheckFreq Frequent, Fine-Grained DNN Checkpointing FAST21
- » Rethinking File Mapping for Persistent Memory FAST21
- » PIDS Attribute Decomposition for Improved Compression and Query Performance in Columnar Storage VLDB20
- » Near-Optimal Latency Versus Cost Tradeoffs in Geo-Distributed Storage NSDI21
- » The Storage Hierarchy is Not a Hierarchy Optimizing Caching on Modern Storage Devices with Orthus FAST21
- » Facebook’s Tectonic Filesystem Efficiency from Exascale FAST21
- » Exploiting Combined Locality for Wide-Stripe Erasure Coding in Distributed Storage FAST21
- » Concordia Distributed Shared Memory with In-Network Cache Coherence FAST21
- » SpanDB A Fast, Cost-Effective LSM-tree Based KV Store on Hybrid Storage FAST21
- » Cerebro A Data System for Optimized Deep Learning Model Selection VLDB20
- » Analyzing and Mitigating Data Stalls in DNN Training VLDB21
- » How to Copy Files FAST20
- » DistCache Provable Load Balancing for LargeScale Storage Systems with Distributed Caching FAST19
- » FIRM An Intelligent Fine-grained Resource Management Framework for SLO-Oriented Microservices OSDI20
- » PACEMAKER Avoiding HeART attacks in storage clusters with disk-adaptive redundancy OSDI20
- » LinnOS Predictability on Unpredictable Flash Storage with a Light Neural Network OSDI20
- » Sundial Fault-tolerant Clock Synchronization for Datacenters OSDI20
- » HetPipe Enabling Large DNN Training on (Whimpy) Heterogeneous GPU Clusters through Integration of Pipelined Model Parallelism and Data Parallelism ATC20
- » Fast RDMA-based Ordered Key-Value Store using Remote Learned Cache OSDI20
- » OpenEC Toward Unified and Configurable Erasure Coding Management in Distributed Storage Systems FAST19
- » A Unified Architecture for Accelerating Distributed DNN Training in Heterogeneous GPU/CPU Clusters OSDI20
- » An Empirical Guide to the Behavior and Use of Scalable Persistent Memory FAST20
- » BASTION A Security Enforcement Network Stack for Container Networks ATC20
- » GIFT A Coupon Based Throttle-and-Reward Mechanism for Fair and Efficient I/O Bandwidth Management on Parallel Storage Systems FAST20
- » MatrixKV Reducing Write Stalls and Write Amplification in LSM-tree Based KV Stores with a Matrix Container in NVM ATC20
container network
persistent memory
NVM
- » Enabling High-Performance and Secure Userspace NVM File Systems with the TRIO Architecture SOSP23
- » Light-Dedup A Light-weight Inline Deduplication Framework for Non-Volatile Memory File Systems ATC23
- » ResPCT Fast Checkpointing in Non-volatile Memory for Multi-threaded Applications Eurosys22
- » Witcher Systematic Crash Consistency Testing for Non-Volatile Memory Key-Value Stores SOSP21
- » WineFS a hugepage-aware file system for persistent memory that ages gracefully ATC21
- » Characterizing and Optimizing Remote Persistent Memory with RDMA and NVM ATC21
- » FlashNeuron SSD-Enabled Large-Batch Training of Very Deep Neural Networks FAST21
- » Exploring the Design Space of Page Management for Multi-Tiered Memory Systems ATC21
- » Optimizing Storage Performance with Calibrated Interrupts FAST21
- » An Empirical Guide to the Behavior and Use of Scalable Persistent Memory FAST20
NVSL lab
SSD
- » NVMe SSD Failures in the Field the Fail-Stop and the Fail-Slow ATC22
- » p2KVS a Portable 2-Dimensional Parallelizing Framework to Improve Scalability of Key-value Stores on SSDs EuroSys22
- » Operational Characteristics of SSDs in Enterprise Storage Systems A Large-Scale Field Study FAST22
- » Improving the Reliability of Next Generation SSDs using WOM-v Codes FAST22
- » Closing the B+ -tree vs LSM-tree Write Amplification Gap on Modern Storage Hardware with Built-in Transparent Compression FAST22
- » Tiny-Tail Flash Near-Perfect Elimination of Garbage Collection Tail Latencies in NAND SSDs FAST17
- » FragPicker A New Defragmentation Tool for Modern Storage Devices SOSP21
- » Austere Flash Caching with Deduplication and Compression ATC20
Erasure Code
- » ELECT Enabling Erasure Coding Tiering for LSM-tree-based Storage FAST24
- » Explore Data Placement Algorithm for Balanced Recovery Load Distribution ATC23
- » ParaRC Embracing Sub-Packetization for Repair Parallelization in MSR-Coded Storage FAST23
- » Hydra Resilient and Highly Available Remote Memory FAST22
- » 宽列纠删码 nan
- » Geometric Partitioning Explore the Boundary of Optimal Erasure Code Repair SOSP21
- » LogECMem Coupling Erasure-Coded In-Memory Key-Value Stores with Parity Logging SC21
- » Boosting Full-Node Repair in Erasure-Coded Storage ATC21
- » OpenEC Toward Unified and Configurable Erasure Coding Management in Distributed Storage Systems FAST19
Pipeline NN
data structure
- » SpanDB A Fast, Cost-Effective LSM-tree Based KV Store on Hybrid Storage FAST21
- » From WiscKey to Bourbon- A Learned Index for Log-Structured Merge Trees OSDI20
transaction system
time synchronize
micro service
- » FIRM An Intelligent Fine-grained Resource Management Framework for SLO-Oriented Microservices OSDI20
machine learning
- » FIRM An Intelligent Fine-grained Resource Management Framework for SLO-Oriented Microservices OSDI20
cache
- » The Storage Hierarchy is Not a Hierarchy Optimizing Caching on Modern Storage Devices with Orthus FAST21
- » Concordia Distributed Shared Memory with In-Network Cache Coherence FAST21
- » DistCache Provable Load Balancing for LargeScale Storage Systems with Distributed Caching FAST19
file system
- » Rethinking File Mapping for Persistent Memory FAST21
- » How to Copy Files FAST20
RDMA
- » Hydra Resilient and Highly Available Remote Memory FAST22
- » RDMA is Turing complete, we just did not know it yet! NSDI22
- » Microsecond Consensus for Microsecond Applications OSDI20
replication system
DNN
- » Looking Beyond GPUs for DNN Scheduling on Multi-Tenant Clusters OSDI22
- » CheckFreq Frequent, Fine-Grained DNN Checkpointing FAST21
- » Analyzing and Mitigating Data Stalls in DNN Training VLDB21
optimization
Deep Learning
model selection
erasure code
- » Clay Codes Moulding MDS Codes to Yield an MSR Code FAST18
- » Exploiting Combined Locality for Wide-Stripe Erasure Coding in Distributed Storage FAST21
wide stripe
storage system
latency
PM
- » TreeSLS A Whole-system Persistent Microkernel with Tree-structured State Checkpoint on NVM SOSP23
- » Pacman An Efficient Compaction Approach for Log-Structured Key-Value Store on Persistent Memory ATC22
- » Bridging the Performance Gap for Copy-based Garbage Collectors atop Non-Volatile Memory EuroSys21
- » ROART Range-query Optimized Persistent ART FAST21
- » Rethinking File Mapping for Persistent Memory FAST21
checkpoint
GPU
- » Gandiva Introspective Cluster Scheduling for Deep Learning OSDI18
- » CheckFreq Frequent, Fine-Grained DNN Checkpointing FAST21
LSM tree；
Data Structure
Operating system
Rack
Schedule
edge calculation
- » Achieving Low Tail-latency and High Scalability for Serializable Transactions in Edge Computing Eurosys21
logging
linux kernel
block layer
scheduler
Federated Learning
log
- » TencentCLS The Cloud Log Service with High Query Performances VLDB22
- » LogGrep+ improve retrieval efficiency on highly compressed cloud log with encoding aware schemes nan
- » NanoLog A Nanosecond Scale Logging System ATC18
- » CLP Efficient and Scalable Search on Compressed Text Logs OSDI21
log search
GC
- » MiDAS Minimizing Write Amplification in Log-Structured Systems through Adaptive Group Number and Size Configuration FAST24
- » Bridging the Performance Gap for Copy-based Garbage Collectors atop Non-Volatile Memory EuroSys21
Java
ML
- » Gemini Fast Failure Recovery in Distributed Training with In-Memory Checkpoints SOSP23
- » Cachew ML input Data Processing as a Service ATC22
- » 存储盘用户画像刻画 nan
- » TVStore Automatically Bounding Time Series Storage via Time-Varying Compression FAST22
- » DeepSketch A New Machine Learning-Based Reference Search Technique for Post-Deduplication Delta Compression FAST22
- » FlashNeuron SSD-Enabled Large-Batch Training of Very Deep Neural Networks FAST21
- » Privacy Budget Scheduling OSDI21
- » Jump-Starting Multivariate Time Series Anomaly Detection for Online Service Systems ATC21
privacy
- » Privacy Budget Scheduling OSDI21
LSM tree
- » ELECT Enabling Erasure Coding Tiering for LSM-tree-based Storage FAST24
- » p2KVS a Portable 2-Dimensional Parallelizing Framework to Improve Scalability of Key-value Stores on SSDs EuroSys22
- » Differentiated Key-Value Storage Management for Balanced I/O Performance ATC21
Transaction
System
- » WineFS a hugepage-aware file system for persistent memory that ages gracefully ATC21
- » Optimizing Storage Performance with Calibrated Interrupts FAST21
Cache
- » Symbiosis The Art of Application and Kernel Cache Cooperation FAST24
- » Take Out the TraChe Maximizing (Tra)nsactional Ca(che) Hit Rate OSDI23
- » Cachew ML input Data Processing as a Service ATC22
- » GL-Cache Group-level learning for efficient and high-performance caching FAST23
- » TriCache A User-Transparent Block Enabling High-Performance Out-of-Core Processing with In-Memory Programs OSDI22
- » OSCA An Online-Model Based Cache Allocation Scheme in Cloud Block Storage Systems ATC20
OS
fragment
DNA Storage
- » What is DNA Storage nan
IMR Disk
Liberation
AI
- » Orca A Distributed Serving System for Transformer-Based Generative Models OSDI22
- » Zeus Locality-aware Distributed Transactions Eurosys21
- » Microsecond-scale Preemption for Concurrent GPU-accelerated DNN Inferences OSDI22
- » iVPF Numerical Invertible Volume Preserving Flow for Efficient Lossless Compression CVPR21
- » Tsunami A Learned Multi-dimensional Index for Correlated Data and Skewed Workloads VLDB21
Crash Consistency
File System
- » Enabling High-Performance and Secure Userspace NVM File Systems with the TRIO Architecture SOSP23
- » Fisc A Large-scale Cloud-native-oriented File System FAST23
- » CFS Scaling Metadata Service for Distributed File System via Pruned Scope of Critical Sections Eurosys23
- » InfiniFS An Efficient Metadata Service for Large-Scale Distributed Filesystems FAST22
- » Modernizing File System through In-Storage Indexing OSDI21
KV
- » Pacman An Efficient Compaction Approach for Log-Structured Key-Value Store on Persistent Memory ATC22
- » DEPART Replica Decoupling for Distributed Key-Value Storage FAST22
- » Modernizing File System through In-Storage Indexing OSDI21
Turing Complete
compression
- » Elf Erasing-based Lossless Floating-Point Compression VLDB23
- » Language Model is Compression Arxiv23
- » LogGrep+ improve retrieval efficiency on highly compressed cloud log with encoding aware schemes nan
- » MorphStore Analytical Query Engine with a Holistic Compression-Enabled Processing Model VLDB20
- » Adaptive Compression for Fast scans on string columns SIGMOD21
- » iVPF Numerical Invertible Volume Preserving Flow for Efficient Lossless Compression CVPR21
Compression
- » BTRBLOCKS Efficient Columnar Compression for Data Lakes SIGMOD23
- » MedCompressor：医学图像压缩 科研分析
- » Chimp Efficient Lossless Floating Point Compression for Time Series Databases VLDB22
- » Hi-Speed DNN Training with Espresso Unleashing the Full Potential of Gradient Compression with Near-Optimal Usage Strategies Eurosys23
- » Neural Compression Review nan
- » CompressDB Enabling Efficient Compressed Data Direct Processing for Various Databases SIGMOD22
- » LogGrep Eurosys21
- » TVStore Automatically Bounding Time Series Storage via Time-Varying Compression FAST22
- » JSON Tiles Fast Analytics on Semi-Structured Data SIGMOD21
- » DeepSketch A New Machine Learning-Based Reference Search Technique for Post-Deduplication Delta Compression FAST22
EC
Compile
Study
- » ADOC Automatically Harmonizing Dataflow Between Components in Log-Structured Key-Value Stores for Improved Performance FAST23
- » NVMe SSD Failures in the Field the Fail-Stop and the Fail-Slow ATC22
- » Operational Characteristics of SSDs in Enterprise Storage Systems A Large-Scale Field Study FAST22
FS
metadata
Storage
- » 存储盘用户画像刻画 nan
storage
- » Where did my 256 GB go? A Measurement Analysis of Storage Consumption on Smart Mobile Devices SIGMETRIC21
study paper
- » Metastable Failures in the Wild OSDI22
- » Where did my 256 GB go? A Measurement Analysis of Storage Consumption on Smart Mobile Devices SIGMETRIC21
database
Persistent Memory
NVMe
Fault Tolerance
- » Gemini Fast Failure Recovery in Distributed Training with In-Memory Checkpoints SOSP23
- » Demystifying and Checking Silent Semantic Violations in Large Distributed Systems OSDI22
- » Carbink Fault-Tolerant Far Memory OSDI22
- » Geometric Partitioning Explore the Boundary of Optimal Erasure Code Repair SOSP21
- » NVMe SSD Failures in the Field the Fail-Stop and the Fail-Slow ATC22
FTL
- » BlockFlex Enabling Storage Harvesting with Software-Defined Flash in Modern Cloud Platforms OSDI22
Open channel SSD
- » BlockFlex Enabling Storage Harvesting with Software-Defined Flash in Modern Cloud Platforms OSDI22
CXL
transaction
- » Zeus Locality-aware Distributed Transactions Eurosys21
Wide Stripe EC
- » 宽列纠删码 nan
DB
Erasure Coding
- » Design Considerations and Analysis of Multi-Level Erasure Coding in Large-Scale Data Centers SC23
- » Tiger Disk-Adaptive Redundancy Without Placement Restrictions OSDI22
eBPF
Checkpoint
study
JIT
Tracing
Debugging
RAID
- » eZNS An Elastic Zoned Namespace for Commodity ZNS SSDs OSDI23
- » Zebra ZRWA-EnaBled Redundant Array of Zoned Namespace SSDs 科研分享
- » RAIZN Redundant Array of Independent Zoned Namespaces ASPLOS21
- » StRAID Stripe-threaded Architecture for Parity-based RAIDs with Ultra-fast SSDs ATC22
search engine
Machine Learning
- » Efficient Memory Management for Large Language Model Serving with PagedAttention SOSP23
- » GL-Cache Group-level learning for efficient and high-performance caching FAST23
- » Hi-Speed DNN Training with Espresso Unleashing the Full Potential of Gradient Compression with Near-Optimal Usage Strategies Eurosys23
Clay Code
KV Store
Fail-Slow
ZNS
SDC
- » SDC Study 科研分享
CPU
- » SDC Study 科研分享
Smart Phone
Deduplication
- » Light-Dedup A Light-weight Inline Deduplication Framework for Non-Volatile Memory File Systems ATC23
- » TiDedup A New Distributed Deduplication Architecture for Ceph ATC23
- » InftyDedup Scalable and Cost-Effective Cloud Tiering with Deduplication FAST23
Metadata
- » CFS Scaling Metadata Service for Distributed File System via Pruned Scope of Critical Sections Eurosys23
HPC
- » HadaFS A File System Bridging the Local and Shared Burst Buffer for Exascale Supercomputers FAST23
Burst Buffer
- » HadaFS A File System Bridging the Local and Shared Burst Buffer for Exascale Supercomputers FAST23
SMR
Serverless
Virtual Machine
Reliability
- » Hyrax Fail-in-Place Server Operation in Cloud Platforms OSDI23
- » Multi-view Feature-based SSD Failure Prediction What, When, and Why FAST23
Failure Prediction
Data Management
Kernel
Page Allocation
ZNS SSD
- » eZNS An Elastic Zoned Namespace for Commodity ZNS SSDs OSDI23
- » Zebra ZRWA-EnaBled Redundant Array of Zoned Namespace SSDs 科研分享
LLM
- » Language Model is Compression Arxiv23
- » Efficient Memory Management for Large Language Model Serving with PagedAttention SOSP23