A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf
.
Filters
Append is Near: Log-based Data Management on ZNS SSDs
2022
Conference on Innovative Data Systems Research
To eliminate these disadvantages, Zoned Namespace (ZNS) SSDs have recently been introduced. ...
We explore advantages of ZNS SSDs with Zone Append, Group Append, and computational storage in four log-based data management areas: (i) log-based file systems, (ii) LSM trees such as RocksDB, (iii) database ...
We thank Pat Helland from Salesforce for the feedback we received from him. We are grateful for the feedback and support we received from our collaborators at the CRSS, Ethan L. Miller, Darrell D. E. ...
dblp:conf/cidr/PurandareWLF22
fatcat:qiura3ovijhnzbc5icmbggdegy
The Design and Implementation of a High-Performance Log-Structured RAID System for ZNS SSDs
[article]
2024
arXiv
pre-print
Zoned Namespace (ZNS) defines a new abstraction for host software to flexibly manage storage in flash-based SSDs as append-only zones. ...
We propose ZapRAID, a high-performance log-structured RAID system for ZNS SSDs by carefully exploiting Zone Append to achieve high write parallelism and lightweight stripe management. ...
Storage systems for ZNS SSDs. Several studies explore new storage system designs for ZNS SSDs. Bjorling et al. ...
arXiv:2402.17963v1
fatcat:ixcjemdnenh4nmivnm4k5bseey
Efficient LSM-Tree Key-Value Data Management on Hybrid SSD/HDD Zoned Storage
[article]
2022
arXiv
pre-print
Zoned storage devices, such as zoned namespace (ZNS) solid-state drives (SSDs) and host-managed shingled magnetic recording (HM-SMR) hard-disk drives (HDDs), expose interfaces for host-level applications ...
We make a case for key-value (KV) stores based on log-structured merge trees (LSM-trees) as host-level applications, and present HHZS, a middleware system that bridges an LSM-tree KV store with hybrid ...
Zoned namespace (ZNS) SSDs and host-managed SMR (HM-SMR) HDDs are two mainstream zoned storage devices available today. ...
arXiv:2205.11753v1
fatcat:fefuu3zh3jgt3d46feo4vqffda
SplitZNS: Towards an Efficient LSM-tree on Zoned Namespace SSDs
2023
ACM Transactions on Architecture and Code Optimization (TACO)
The Zoned Namespace (ZNS) Solid State Drive (SSD) is a nascent form of storage device that offers novel prospects for the Log Structured Merge Tree (LSM-tree). ...
In this paper, we present SplitZNS, which introduces small zones by tweaking the zone-to-chip mapping to maximize GC efficiency for LSM-tree on ZNS SSDs. ...
The emerging storage device, Zoned Namespace (ZNS) [5, 16] SSD affords LSM-tree with new opportunities. ...
doi:10.1145/3608476
fatcat:akqxqnvkrjewhjukp3lf4dwdxm
Immutability changes everything
2015
Communications of the ACM
For a relational database, this can be a key-value pair for each record or each index entry. ...
Read Perspiration Data may be copied many times when we use copy on write (e.g. with log structured file systems, log structured merge systems, wear leveling in SSDs, and shingle management in HDD). ...
doi:10.1145/2844112
fatcat:7t5f6gtalvfcxecl7q6zw2mrzi
We quantify the performance benefits of Strata using a 3-layer storage hierarchy of emulated NVM, a flash-based SSD, and a high-density HDD. ...
At its heart, Strata uses a log-structured approach with a novel split of responsibilities among user mode, kernel, and storage layers that separates the concerns of scalable, high-performance persistence ...
For their insights and valuable comments, we thank the anonymous reviewers and our shepherd Ashvin Goel. We acknowledge funding from NSF grants NSF-1518702 and CNS-1618563. ...
doi:10.1145/3132747.3132770
dblp:conf/sosp/KwonFHPWA17
fatcat:zcrqfftt35elrloxnwup6oarje
Real-Time Semantic Search Using Approximate Methodology for Large-Scale Storage Systems
2016
IEEE Transactions on Parallel and Distributed Systems
FAST is further improved by using semantic-aware namespace to provide dynamic and adaptive namespace management for ultra-large storage systems. ...
The idea behind FAST is to explore and exploit the semantic correlation within and among datasets via correlation-aware hashing and manageable flat-structured addressing to significantly reduce the processing ...
Clearly, since the most expensive resource is memory and it is the insufficient memory space for index structures that causes most of the execution delays, spaceefficient index structures are a key to ...
doi:10.1109/tpds.2015.2425399
fatcat:toymxuf5drgl5lnvyefefuaccu
File Systems Fated for Senescence? Nonsense, Says Science!
2017
USENIX Conference on File and Storage Technologies
File systems must allocate space for files without knowing what will be added or removed in the future. ...
present microbenchmarks demonstrating that common placement strategies are extremely sensitive to file-creation order; varying the creation order of a few thousand small files in a real-world directory structure ...
Acknowledgments We thank the anonymous reviewers and our shepherd Philip Shilane for their insightful comments on earlier drafts of the work. ...
dblp:conf/fast/ConwayBJJZYBJKP17
fatcat:iuadgvb7kvbz3npxyqkui7ztyq
Building workload-independent storage with VT-trees
2013
USENIX Conference on File and Storage Technologies
Current storage systems focus on optimizing for one band of workloads at the expense of other workloads due to limitations in existing storage system data structures. ...
Our evaluation shows that our user-level system has 2-6.6× better performance for random-write workloads and only a small average overhead for other workloads. ...
We thank the anonymous USENIX reviewers for their valuable feedback. This work was supported in part by NSF awards CCF-0937833 and CCF-0937854. ...
dblp:conf/fast/ShettySMASZ13
fatcat:pq2ki6uohrelfabkkuy2ucxfee
File System Aging
[article]
2024
arXiv
pre-print
File systems must allocate space for files without knowing what will be added or removed in the future. ...
We present a framework for understanding and predicting aging, and identify the key features of BetrFS that avoid aging. ...
Each zone root is either a single, large file, or a subdirectory of small files. The key for a file or directory is its relative path to its zone root. ...
arXiv:2401.08858v1
fatcat:7qfs6cljyfeyvprw5tpkonayhe
Forensic Issues and Techniques to Improve Security in SSD with Flex Capacity Feature
2021
IEEE Access
INDEX TERMS Forensic, over-provisioning, hidden area, attack model, malware, NAND flash memory, SSD, invalid block. ...
This leads to a discussion of forensic possibilities and countermeasures for SSDs that can change the over-provisioning area. ...
ZONE NAMESPACE In general, when garbage collection is performed, additional write operations occur. Garbage collection increases Write Amplification Factor (WAF) by default. ...
doi:10.1109/access.2021.3136483
fatcat:rikapdprlzezzpncqhzqnfvqtq
An Approach of Automatic SPARQL Generation for BIM Data Extraction
2020
Applied Sciences
Efficient extraction of BIM data is a necessary and vital step for various data analyses and applications, especially in large-scale BIM projects. ...
In this paper, we develop a simple method for automatic SPARQL (SPARQL Protocol and RDF Query Language) query generation to implement effective data extraction. ...
So, we can add the fixed structure in the shortest path to find the GUID value, shown in the red font in Zone II of Figure 12 . ...
doi:10.3390/app10248794
fatcat:jeqz4j46b5ebrf5wtd5tbd3mku
Enabling low tail latency on multicore key-value stores
2020
Proceedings of the VLDB Endowment
To that aim, we argue in favor of techniques such as an asynchronous programming model, message-passing communication, and log-structured storage on modern hardware. ...
Finally, we thank Thomas Willhalm, Otto Bruggeman and Heinrich Teiken for providing the hardware and technical support for the experiments. ...
Acknowledgements We thank the anonymous reviewers for their helpful feedback. ...
doi:10.14778/3384345.3384356
fatcat:o4esmmmrmnbuxky4blarwuxixa
DXRAM's Fault-Tolerance Mechanisms Meet High Speed I/O Devices
[article]
2018
arXiv
pre-print
This is also the first time DXRAM's backup system is evaluated with high speed I/O devices, specifically with 56 GBit/s InfiniBand interconnect and PCI-e SSDs. ...
Furthermore, we discuss the copyset replica distribution to reduce the probability for data loss and the adaptations to the original approach for DXRAM. ...
For instance, all data can be stored on SSD with indexes in RAM or all data can be stored in RAM and optionally on SSD with a configurable replication factor. ...
arXiv:1807.03562v2
fatcat:p4yobou5vjgqrn4lvdumamuelq
A Cache-Based Data Movement Infrastructure for On-demand Scientific Cloud Computing
[chapter]
2019
Lecture Notes in Computer Science
As cloud computing has become the de facto standard for big data processing, there is interest in using a multi-cloud environment that combines public cloud resources with private on-premise infrastructure ...
We thank Amazon and HUAWEI for contributing cloud resources to this research project. ...
Accordingly, the global namespace is provided using the tree-based directory structure and seamlessly integrates into the namespace of local file system. ...
doi:10.1007/978-3-030-18645-6_3
fatcat:owc6xwtlkndfxnghwd63caxuuy
« Previous
Showing results 1 — 15 out of 79 results