12th USENIX Workshop on Hot Topics in Storage and File Systems (HotStorage '20)
2020-07-13..2020-07-14
-- online --
as the organisers put it
The 12th USENIX Workshop on Hot Topics in Storage and File Systems (HotStorage '20) will be co-located with USENIX ATC '20 and will take place July 13–14, 2020.
| title | author | length |
|---|---|---|
| A New LSM-style Garbage Collection Scheme for ZNS SSDs | Gunhee Choi, Kwanghee Lee, Myunghoon Oh, Jongmoo Choi, Jhuyeong Jhin, Yongseok Oh | 16m |
LSM_ZGC claims to cut ZNS GC latency by dividing 1GB zones into 2MB segments and using segment based fine-grained garbage collection, reading both valid and… | ||
| Ultra-Low Latency SSDs' Impact on Overall Energy Efficiency | Bryan Harris, Nihat Altiparmak | 14m |
Empirical study measures an Intel Optane SSD alongside an HDD and two flash SSDs on a 1U Dell using an inline plug power meter and fio workloads and shows… | ||
| Need for a Deeper Cross-Layer Optimization for Dense NAND SSD to Improve Read | Arpith K, K. Gopinath | 14m |
Proposes combining LSB/CSB/MSB pages in a wordline into a single Melded-pages logical page, implemented in a SimpleSSD simulator with HDFS workloads and… | ||
| Say Goodbye to Off-heap Caches! On-heap Caches Using Memory-Mapped I/O | Iacovos G. Kolokasis, Anastasios Papagiannis, Polyvios Pratikakis, Angelos Bilas, Foivos S. Zakkak | 13m |
Spark extension named TeraCache implements a second non-garbage-collected heap backed by memory-mapped fast storage so cached RDDs avoid serialization and the… | ||
| Differentiating Cache Files for Fine-grain Management to Improve Mobile Performance | Yu Liang, Jinheng Li, Xianzhang Chen, Rachata Ausavarungnirun, Riwei Pan, Tei-Wei Kuo, Chun Jason Xue | 14m |
Analysis of two hours of traces across 10 apps (one case: Facebook generates 1.2 GB of cache) shows cache files vary by lifetime, size and read pattern: writes… | ||
| Desperately Seeking ... Optimal Multi-Tier Cache Configurations | Tyler Estro, Pranav Bhandari, Avani Wildani, Erez Zadok | 15m |
Extends PyMimircache into an n‑level I/O simulator with write‑through/write‑back support, trace upscaling and reverse miniature sampling, and stores experiment… | ||
| Reinforcement Learning-Based SLC Cache Technique for Enhancing SSD Write | Sangjin Yoo, Dongkun Shin | 16m |
Claims a Q-learning agent that monitors workload and internal SSD state and adjusts SLC cache size and hot/cold separation threshold to maximize write… | ||
| SelectiveEC: Selective Reconstruction in Erasure-coded Storage Systems | Liangliang Xu, Min Lyu, Qiliang Li, Lingjiang Xie, Yinlong Xu | 15m |
SelectiveEC is a balanced scheduling module that picks which erasure-coded stripes to rebuild and which source and replacement nodes to use, builds bipartite… | ||
| Rethinking WOM Codes to Enhance the Lifetime in New SSD Generations | Shehbaz Jaffer, Kaveh Mahdaviani, Bianca Schroeder | 14m |
Presents a new, simple family of WOM codes designed for multi-level cells and evaluated on QLC drives, implemented with practical storage overheads and… | ||
| StripeFinder: Erasure Coding of Small Objects Over Key-Value Storage Devices | Umesh Maheshwari | 16m |
StripeFinder exposes that KVMD stores DxP metadata objects and quantifies the cost: for P=2 and 16-byte keys, KVMD gives byte amplification 2.8x (D=4) and 3.3x… | ||
| Prefetching in Hybrid Main Memory Systems | Subisha V, Varun Gohil, Nisarg Ujjainkar, Manu Awasthi | 13m |
Builds a DRAM cache prefetcher for hybrid DRAM+PCM main memory that extends the Alloy Cache design to prefetch and cache at cacheline and page granularities… | ||
| It's Time to Revisit LRU vs. FIFO | Ohad Eytan, Danny Harnik, Effi Ofer, Roy Friedman, Ronen Kat | 13m |
Reevaluates LRU vs FIFO by modeling metadata access cost and simulating caches on 99 cloud-object traces plus standard MSR/TPCC workloads, varying cache sizes… | ||
| Processing in Storage Class Memory | Joel Nider, Craig Mustard, Andrada Zoltan, Alexandra Fedorova | 15m |
Position: processing-in-memory becomes integral to storage-class memory to cut memory-bandwidth and CPU usage by running simple, parallel tasks inside SCM… | ||
| Platform-Agnostic Lightweight Deep Learning for Garbage Collection... | Junhyeok Jang, Donghyun Gouk, Jinwoo Shin, Myoungsoo Jung | 05m |
A platform-agnostic GC scheduler called GC Tutor uses a lightweight DNN classifier to predict the idle time between the last and next I/O request and schedules… | ||
| GramFS: The Graph Model-based Namespace Management of Large-scale... | Hongbo Li, Yubo Liu, Zhiguang Chen, Nong Xiao | 05m |
Claims a graph-structured namespace that replaces hierarchical inodes with nodes and edges to capture directories, hard links and snapshots, implements compact… | ||
| The Key-Value SSD as a First-Class Citizen in the Operating System | Carl Duffy, Sang-Hoon Kim, Jin-Soo Kim | 06m |
Proposes a concrete OS integration: a pseudo file system that exposes buckets as files, enforces bucket-level permissions and locking, and adds a key-value… | ||
| Synergetic effects of Software and Hardware Parameters on the LSM system. | Jinghuan Yu, Heejin Yoon, Sam H. Noh, Young-ri Choi, Chun Jason Xue | 05m |
Shows that LSM tunables interact with storage type: fio benchmarks and RocksDB runs reveal HDD/SSD/NVMe/persistent memory behave differently — fio benchmark… | ||
| TensorPRAM: Designing a Scalable Heterogeneous Deep Learning Accelerator | Sangwon Lee, Gyuyoung Park, Myoungsoo Jung | 06m |
Design replaces DRAM with byte-addressable PRAM inside an FPGA-based accelerator so model weights and activations sit on nonvolatile memory and avoid… | ||
| Can Microservices Drive a Renaissance in Workload-Aware Storage Management? | Pranav Bhandari, Avani Wildani, Dimitris Skourtis, Vasily Tarasov, Deepavali Bhagwat, Lukas Rupprecht, Ali Anwar | 06m |
Paper evaluates microservices using docker and kubernetes with a per-volume io cache and shows a concrete auto-tuning prototype: a collector and recommender… | ||
| GPUKV: Towards a GPU-Driven Computing on Key-Value SSD | Min-Gyo Jeong, Chang-Gyu Lee, Donggyu Park, Sungyong Park, Youngjae Kim, Jungki Noh, Woosuk Chung, Kyoung Park | 05m |
Describes a system that pushes key-value processing onto GPUs co-located with or attached to key-value SSDs, implementing KV-SSD access paths, CUDA kernels for… | ||
| SGX-SSD: A Policy-based Versioning SSD with Intel SGX | Jinwoo Ahn, Seungjin Lee, Jinhoon Lee, Yungwoo Ko, Donghyun Min, Junghee Lee, Youngjae Kim | 06m |
Implements a policy-driven versioning SSD that moves backup logic into firmware and protects metadata and requests with an Intel SGX enclave, so policy changes… | ||
| On Failure Diagnosis of the Storage Stack | Duo Zhang, Om Rameshwar Gatla, Runzhou Han, Mai Zheng | 06m |
A cross-layer diagnosis platform called X-ray combines a host agent that captures kernel and application events with a dev agent that intercepts device… | ||
| CompoundFS: Compounding I/O Operations in Firmware File Systems | Yujie Ren, Jian Zhang, Sudarsun Kannan | 16m |
Claims a firmware-level file system, CompoundFS, that combines consecutive I/O into single commands and offloads simple data processing to device CPUs… | ||
| Can We Store the Whole World's Data in DNA Storage? | Bingzhe Li, Nae Young Song, Li Ou, David H.C. Du | 15m |
Can We Store the Whole World's Data in DNA Storage?
Bingzhe Li, Nae Young Song, Li Ou, and David H.C. Du, University of Minnesota, Twin Cities
The total… | ||
| Too Many Knobs to Tune? Towards Faster Database Tuning by Pre-selecting Important | Konstantinos Kanellis, Ramnatthan Alagappan, Shivaram Venkataraman | 17m |
Experimental study shows that tuning five knobs out of 155 on Cassandra with YCSB-A reaches about 99% of the best throughput, identified by a random forest… | ||
| Neural Trees: Using Neural Networks as an Alternative to Binary Comparison in | Douglas Santry | 15m |
Neural Trees implement a hierarchy of many tiny neural models that map keys to physical addresses instead of learning the data distribution, claiming up to 40%… | ||
| SplitKV: Splitting IO Paths for Different Sized Key-Value Items with Advanced | Shukai Han, Dejun Jiang, Jin Xiong | 18m |
SplitKV is a prototype key-value store that writes small KV to PM then migrates to SSD and directly writes large KV to SSD, using Intel DC persistent memory… | ||
| In support of workload-aware streaming state management | Vasiliki Kalavri, John Liagouris | 16m |
In support of workload-aware streaming state management
Vasiliki Kalavri, Boston University; John Liagouris, Boston University & Hariri Institute for… | ||
| Re-think Data Management Software Design Upon the Arrival of Storage Hardware with | Ning Zheng, Xubin Chen, Jiangpeng Li, Qi Wu, Yang Liu, Yong Peng, Fei Sun, Hao Zhong, Tong Zhang | 14m |
Argues that transparent compression embedded in drives (an FPGA-based flash controller exposing a standard block device) decouples logical from physical… | ||
| Designing a Storage Software Stack for Accelerators | Shinichi Awamoto, Erich Focht, Michio Honda | 15m |
Built hayagui user-space storage stack that detaches NVMe from Linux, maps device registers and DMA buffers into accelerator memory, and offers three… | ||
| A file system for safely interacting with untrusted USB flash drives | Ke Zhong, Zhihao Jiang, Ke Ma, Sebastian Angel | 16m |
RBFuse isolates USB handling by remap host controller into VM and exports the drive as a mountable VFS via a FUSE VFS client-server, eliminating host USB… | ||
| Understanding and Finding Crash-Consistency Bugs in Parallel File Systems | Jinghan Sun, Chen Wang, Jian Huang, Marc Snir | 12m |
Study finds 34 vulnerabilities across parallel file systems, with 15 and 19 in BeeGFS and OrangeFS and 5 in XD4, including two persistence-order bugs that… | ||
| The Case for Benchmarking Control Operations in Cloud Native Storage | Alex Merenstein, Vasily Tarasov, Ali Anwar, Deepavali Bhagwat, Lukas Rupprecht, Dimitris Skourtis, Erez Zadok | 15m |
Argues that current benchmarks omit storage control operations and demonstrates with a five‑node Kubernetes cluster using gluster, ceph and open ebs that these… | ||
| Could cloud storage be disrupted in the next decade? | Andromachi Chatzieleftheriou, Ioan Stefanovici, Dushyanth Narayanan, Benn Thomsen, Antony Rowstron | 14m |
Argues cloud storage faces likely disruption within a decade because flash and HDD are near maturity, tape shipments and vendors shrink, and cloud scale can… | ||
| MicroMon: A Monitoring Framework for Tackling Distributed Heterogeneity | Baber Khalid, Nolan Rudolph, Ramakrishnan Durairajan, Sudarsun Kannan | 16m |
MicroMon is a fine‑grained monitoring framework that collects micrometrics (page cache, per‑host IO and network queue delays, SSD SMART counters), filters them… | ||
| nothing matches | ||