12 Hits in 3.4 sec

Evaluating persistent memory range indexes

Lucas Lersch, Xiangpeng Hao, Ismail Oukid, Tianzheng Wang, Thomas Willhalm
2019 Proceedings of the VLDB Endowment  
With the recent released Intel Optane DC Persistent Memory, for the first time, this paper provides a comprehensive evaluation of recent persistent index structures.  ...  Persistent memory (PM) is fundamentally changing the way database index structures are built by enabling persistence, high performance, and (near) instant recovery all on the memory bus.  ...  CONCLUSION In this paper, we revisited PM-based range indexes on the newly released Intel Optane DC Persistent Memory.  ... 
doi:10.14778/3372716.3372728 fatcat:vu5pnm7umbelnf36rrmw4t3yii

Revisiting active object stores: Bringing data locality to the limit with NVM

Alex Barcelo, Anna Queralt, Toni Cortes
2021 Future generations computer systems  
Our discussion will focus on the active aspect of the system as well as on the implications of the memory configuration.  ...  The proliferation of non-volatile memory (NVM) technology is blurring the line between system memory (fast and scarce) and block devices (slow and abundant).  ...  This work has been partially supported by Intel® under the BSC-Intel collaboration on 3D XPoint™ Technology.  ... 
doi:10.1016/j.future.2021.10.025 fatcat:uxkvms5qbnbnrnlm56spdh5ewe

Analysis and Optimization of Persistent Memory Index Structures' Write Amplification

Youngjoo Woo, Taesoo Kim, Sungin Jung, Euiseong Seo
2021 IEEE Access  
[1] Intel Optane DC Persistent Memory: Benefit From Greater Capacity, put.  ...  - memory/optane-dc-persistent-memory-brief.html tively, and the Ph.D. degree in computer engineer- [33] Intel Corporation. (2020).  ... 
doi:10.1109/access.2021.3136459 fatcat:w5spdozzkfbztj5qtm4laegm4y


Xian-He Sun, Dong Li, Wen-Guang Chen, Tao Li, Ji-Wu Shu, Bo Wu, Jin Xiong, Jinging Xue, Feng Zhang, Ji-Dong Zhai, Zhiia Zhao
2021 Journal of Computer Science and Technology  
His research interests include data-intensive high-performance computing, memory and I/O systems, software system for big data applications, and performance evaluation and optimization.  ...  He is a review board member of IEEE Transaction on Parallel and Distributed Systems (TPDS).  ...  This paper outlines some of the basic performance characteristics of Intel Optane DC persistent memory module (PMM), and explains how it can be configured and used to address the needs of memory and I/  ... 
doi:10.1007/s11390-021-0001-4 fatcat:f6kzf7226jbltpygejrg5wwut4

An Empirical Guide to the Behavior and Use of Scalable Persistent Memory [article]

Jian Yang, Juno Kim, Morteza Hoseinzadeh, Joseph Izraelevitz, Steven Swanson
2019 arXiv   pre-print
Now that 3D XPoint NVDIMMs are actually here, we can provide detailed performance numbers, concrete guidance for programmers on these systems, reevaluate prior art for performance, and reoptimize persistent  ...  With our improved understanding, we then explore the performance of prior art in application-level software for persistent memory, taking note of where their performance was influenced by our guidelines  ...  It implements various index data structures (called "storage engines") to index programs data and uses the Persistent Memory Development Kit (PMDK [14] ) to manage its persistent data.  ... 
arXiv:1908.03583v1 fatcat:vdni6qh2wnhwfo7mfzmudv7faa

Enabling low tail latency on multicore key-value stores

Lucas Lersch, Ivan Schreter, Ismail Oukid, Wolfgang Lehner
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.  ...  However, most KVS have similar design decisions which focus on improving throughput metrics, at times by sacrificing latency.  ...  We also thank Badrish Chandramouli, Ryan Stutsman and Chinmay Kulkarni for their help in tuning FASTER and comments on benchmark results.  ... 
doi:10.14778/3384345.3384356 fatcat:o4esmmmrmnbuxky4blarwuxixa

Kernel/User-level Collaborative Persistent Memory File System with Efficiency and Protection [article]

Youmin Chen, Youyou Lu, Bohong Zhu, Jiwu Shu
2019 arXiv   pre-print
Experiments on Optane DC persistent memory show that KucoFS significantly outperforms existing file systems and shows better scalability.  ...  Emerging high performance non-volatile memories recall the importance of efficient file system design.  ...  Introduction Emerging byte-addressable non-volatile memories (NVMs), such as PCM [19, 26, 38] , ReRAM [5] , and the recently released Intel Optane DC persistent memory [4] , provide performance comparable  ... 
arXiv:1908.10740v1 fatcat:n4cjzyktuvg2tezmxvj3clgogi

Efficient Kernel Object Management for Tiered Memory Systems with KLOC [article]

Sudarsun Kannan, Yujie Ren, Abhishek Bhatacharjee
2020 arXiv   pre-print
Unfortunately, modern OSes do not support efficient tiering of data between heterogeneous memories.  ...  Delivering on this promise requires an efficient operating system (OS) mechanisms and policies for data management.  ...  KLOC performance on DC-Optane Memory Finally, to understand the impact of KLOC on DC-Optane memory technologies, we use a 256GB DC-Optane on a memory socket with 16GB DRAM hardware managed cache (memory  ... 
arXiv:2004.04760v1 fatcat:23baz7oi6vdgtatw5ltuok4rne

Eurolab-4-HPC Long-Term Vision on High-Performance Computing [article]

Theo Ungerer, Paul Carpenter
2018 arXiv   pre-print
The proposal on research topics is derived from the report and discussions within the road mapping expert group.  ...  This document presents the "EuroLab-4-HPC Long-Term Vision on High-Performance Computing" of August 2017, a road mapping effort within the EC CSA1 Eurolab-4-HPC that targets potential changes in hardware  ...  Intel/Micro 3D-Xpoint memory is now available as Optane-SSDs DC HBM DRAM Die HBM DRAM Die HBM DRAM Die HBM DRAM Die TSV Microbump Logic Die PHY GPU/CPU/SoC Die PHY Interposer Package  ... 
arXiv:1807.04521v1 fatcat:5neetrgubjhnvcajcktpkohrzq

Sage: Parallel Semi-Asymmetric Graph Algorithms for NVRAMs [article]

Laxman Dhulipala, Charlie McGuffey, Hongbo Kang, Yan Gu, Guy E. Blelloch, Phillip B. Gibbons, Julian Shun
2020 arXiv   pre-print
DC Persistent Memory, and show that Sage outperforms the fastest prior systems designed for NVRAM.  ...  We experimentally study Sage using a 48-core machine on the largest publicly-available real-world graph (the Hyperlink Web graph with over 3.5 billion vertices and 128 billion edges) equipped with Optane  ...  the market (e.g., Intel's Optane DC Persistent Memory).  ... 
arXiv:1910.12310v2 fatcat:kba4cfosrzgcbf3e2zzatvim7i

Performance evaluation and comparison of NVMe memory and Lustre File System on CLAIX

Lukas Aldenhoff, Matthias Stefan Müller, Uwe Naumann
) Table 4 . 4 1: Options for IOR that were used in the thesis The used NVMe SSDs are from Intels SSD DC P3600 series and are also plugged in on CLAIX-MPI nodes.  ...  Read 2600 MB/s Latency 20 s Random Read 450.000 IOPS Random Write 56.000 IOPS Interface PCIe NVMe 3.0 x4 (b) Intel R ○ SSD DC P3600 specifications [?]  ...  on how to take advantage of the system.  ... 
doi:10.18154/rwth-2018-223515 fatcat:4ovmjpniwnbntgjusxhpidnjua

Efficient in-hardware compression of on-chip data

Amin Ghasemazar
In order to efficiently compress on-chip data, the compression mechanisms need to leverage all relevant data stored in on-chip memory.  ...  On-chip memory capacities, however, have not kept up: average CPU last-level cache capacity per core (thread) has stagnated at 1MB.  ...  Large-scale in-memory analytics on intel® optane™ dc persistent memory.  ... 
doi:10.14288/1.0404515 fatcat:nxtj5xz4yffm7inyjyp7k6caem