Filters








7,809 Hits in 4.3 sec

CPU and cache efficient management of memory-resident databases

H. Pirk, F. Funke, M. Grund, T. Neumann, U. Leser, S. Manegold, A. Kemper, M. Kersten
2013 2013 IEEE 29th International Conference on Data Engineering (ICDE)  
Memory-Resident Database Management Systems (MRDBMS) have to be optimized for two resources: CPU cycles and memory bandwidth.  ...  . • We present the design and implementation of a PDSMbased storage component for HyPer, our JiT-based Database Management System (DBMS).  ...  RELATED WORK Before illustrating our approach to CPU and cache efficient memory-resident data management, we discuss previous approaches, the encountered problems and inherent tradeoffs. A.  ... 
doi:10.1109/icde.2013.6544810 dblp:conf/icde/PirkFG0LMKK13 fatcat:k5fzaue25ffhhpc4hjozcx2pji

Performance evaluation of extended storage architectures for transaction processing

Erhard Rahm
1992 Proceedings of the 1992 ACM SIGMOD international conference on Management of data - SIGMOD '92  
The buffer manager (BM) is responsible for caching of data- base pages in main memory, for logging and for managing a write buffer and/or database cache in extended memory (NVEM).  ...  A more space-efficient usage of the new memory types results if they are used as a write buffer or for caching database pages at an additional level of the storage hierarchy.A write buffer permits log  ... 
doi:10.1145/130283.130329 dblp:conf/sigmod/Rahm92 fatcat:cyhe74xy2ngmponm5a7wcwr5ue

Analyzing the energy efficiency of a database server

Dimitris Tsirogiannis, Stavros Harizopoulos, Mehul A. Shah
2010 Proceedings of the 2010 international conference on Management of data - SIGMOD '10  
We then experiment with several classes of database systems and storage managers, varying parameters that span from different query plans to compression algorithms and from physical layout to CPU frequency  ...  In this paper, we focus on the role of database software in affecting, and, ultimately, improving the energy efficiency of a server.  ...  Kozyrakis, and Partha Ranganathan.  ... 
doi:10.1145/1807167.1807194 dblp:conf/sigmod/TsirogiannisHS10 fatcat:ddermari6nhezaxawigt7lkf4a

In-memory preprocessing of streaming sensory data - a partitioned relational database approach

József Marton, Sándor Gajdos
2010 Periodica Polytechnica Electrical Engineering  
As sensory data stream in the database it is efficient to preprocess them in-memory before integrating in the repository in order to save storage I/O cost.  ...  In this paper we present a database architecture and an application area and method in detail.  ...  IBM's soliDB in-memory relational database management system can also operate as a cache over a disk-resident database, called soliDB Universal Cache.  ... 
doi:10.3311/pp.ee.2010-3-4.06 fatcat:gujiuyxk4betzmeqymwqdlehw4

GPUfs

Mark Silberstein, Bryan Ford, Idit Keidar, Emmett Witchel
2013 Proceedings of the eighteenth international conference on Architectural support for programming languages and operating systems - ASPLOS '13  
GPUfs provides a POSIX-like API for GPU programs, exploits GPU parallelism for efficiency, and optimizes GPU file access by extending the host CPU's buffer cache into GPU memory.  ...  Our experiments, based on a set of real benchmarks adapted to use our file system, demonstrate the feasibility and benefits of the GPUfs approach.  ...  Acknowledgments This research was supported in part by NSF grants CNS-1017785 and CNS-1017206, by the Andrew and Erna Fince Viterbi Fellowship, and by a 2010 NVIDIA research award.  ... 
doi:10.1145/2451116.2451169 dblp:conf/asplos/SilbersteinFKW13 fatcat:2dt63vkb4jbkjek66ascxvhrb4

GPUfs

Mark Silberstein, Bryan Ford, Idit Keidar, Emmett Witchel
2014 ACM Transactions on Computer Systems  
GPUfs provides a POSIX-like API for GPU programs, exploits GPU parallelism for efficiency, and optimizes GPU file access by extending the host CPU's buffer cache into GPU memory.  ...  Our experiments, based on a set of real benchmarks adapted to use our file system, demonstrate the feasibility and benefits of the GPUfs approach.  ...  Acknowledgments This research was supported in part by NSF grants CNS-1017785 and CNS-1017206, by the Andrew and Erna Fince Viterbi Fellowship, and by a 2010 NVIDIA research award.  ... 
doi:10.1145/2553081 fatcat:xheymqfpm5cofkdejr2ulmkp3q

GPUfs

Mark Silberstein, Bryan Ford, Idit Keidar, Emmett Witchel
2013 SIGPLAN notices  
GPUfs provides a POSIX-like API for GPU programs, exploits GPU parallelism for efficiency, and optimizes GPU file access by extending the host CPU's buffer cache into GPU memory.  ...  Our experiments, based on a set of real benchmarks adapted to use our file system, demonstrate the feasibility and benefits of the GPUfs approach.  ...  Acknowledgments This research was supported in part by NSF grants CNS-1017785 and CNS-1017206, by the Andrew and Erna Fince Viterbi Fellowship, and by a 2010 NVIDIA research award.  ... 
doi:10.1145/2499368.2451169 fatcat:toc5gnurjfhbvofoybnvt5ms5e

Exploiting Coarse-Grained Parallelism in B+ Tree Searches on an APU

Mayank Daga, Mark Nutter
2012 2012 SC Companion: High Performance Computing, Networking Storage and Analysis  
We also present an analysis of the effect of caches on performance, and of the efficacy of the APU to eliminate data-copies.  ...  B+ tree structured index searches are one of the fundamental database operations and hence, accelerating them is essential.  ...  ACKNOWLEDGMENTS We thank Jay Cornwall and Sean Keely for helping with the technical aspects, Takahiro Harada and Lee Howes for providing the radix sort implementation, and Dinesh Manocha and Vinod Tipparaju  ... 
doi:10.1109/sc.companion.2012.40 dblp:conf/sc/DagaN12 fatcat:xudqsapmi5an7il2adpxsn4pda

In-Memory Big Data Management and Processing: A Survey

Hao Zhang, Gang Chen, Beng Chin Ooi, Kian-Lee Tan, Meihui Zhang
2015 IEEE Transactions on Knowledge and Data Engineering  
Many of these researches have focused along several dimensions: modern CPU and memory hierarchy utilization, time/space efficiency, parallelism and concurrency control.  ...  We also give a comprehensive presentation of important technology in memory management, and some key factors that need to be considered in order to achieve efficient in-memory data management and processing  ...  We would like to thank the anonymous reviewers, and also Bingsheng He, Eric Lo and Bogdan Marius Tudor, for their insightful comments and suggestions.  ... 
doi:10.1109/tkde.2015.2427795 fatcat:u7r3rtvhxbainfeazfduxcdwrm

Hetero-DB: Next Generation High-Performance Database Systems by Best Utilizing Heterogeneous Computing and Storage Resources

Kai Zhang, Feng Chen, Xiaoning Ding, Yin Huai, Rubao Lee, Tian Luo, Kaibo Wang, Yuan Yuan, Xiaodong Zhang
2015 Journal of Computer Science and Technology  
Hetero-DB develops a GPU-aware query execution engine with GPU device memory management and query scheduling mechanism to support concurrent query execution.  ...  To best utilize the hybrid hardware devices, the semantic information that is critical for storage I/O is identified and passed to the storage manager, which has a great potential to improve the efficiency  ...  There are three main differences between data replacement in device memory and conventional CPU buffer pool management.  ... 
doi:10.1007/s11390-015-1553-y fatcat:tv5xc6p4avhebfkhmzj5emcl2y

Improving In-Memory Database Operations with Acceleration DIMM (AxDIMM)

Donghun Lee, Jinin So, MINSEON AHN, Jong-Geon Lee, Jungmin Kim, Jeonghyeon Cho, Rebholz Oliver, Vishnu Charan Thummala, Ravi shankar JV, Sachin Suresh Upadhya, Mohammed Ibrahim Khan, Jin Hyun Kim
2022 Data Management on New Hardware  
The significant overhead needed to transfer the data between CPUs and memory devices is one of the hottest issues in many areas of computing, such as database management systems.  ...  With the minimized data transfer overhead, it reduces CPU resource consumption, relieves the memory bandwidth bottleneck, and boosts energy efficiency.  ...  host CPUs and DBA engines and cache coherency.  ... 
doi:10.1145/3533737.3535093 fatcat:7tnrney23ff7bmnanar32roewq

"Anti-Caching"-based elastic memory management for Big Data

Hao Zhang, Gang Chen, Beng Chin Ooi, Weng-Fai Wong, Shensen Wu, Yubin Xia
2015 2015 IEEE 31st International Conference on Data Engineering  
, in terms of memory and CPU overhead, and memory shortage scenarios.  ...  higher memory and CPU overhead.  ... 
doi:10.1109/icde.2015.7113375 dblp:conf/icde/ZhangCOWWX15 fatcat:kbvzzu7onbfylpsxn6p2ul74pi

GPU-Accelerated Database Systems: Survey and Open Challenges [chapter]

Sebastian Breß, Max Heimel, Norbert Siegmund, Ladjel Bellatreche, Gunter Saake
2014 Lecture Notes in Computer Science  
In the past years, there were many approaches to make use of GPUs at different levels of a database system. In this paper, we explore the design space of GPU-accelerated database management systems.  ...  The vast amount of processing power and memory bandwidth provided by modern graphics cards make them an interesting platform for data-intensive applications.  ...  Jedox AG and the anonymous reviewers of the GPUs in Databases Workshop for their helpful feedback on the workshop version of this paper [17] .  ... 
doi:10.1007/978-3-662-45761-0_1 fatcat:rpwqxejbkjh6dhzp27ppiawzsu

GAIA: An OS Page Cache for Heterogeneous Systems

Tanya Brokhman, Pavel Lifshits, Mark Silberstein
2019 USENIX Annual Technical Conference  
We propose a principled approach to integrating GPU memory with an OS page cache. We design GAIA, a weakly-consistent page cache that spans CPU and GPU memories.  ...  performance by using data cached in GPU memory, and (3) optimizes the readahead prefetcher to support accesses to files cached in GPUs.  ...  We also gratefully acknowledge the support of the Israel Science Foundation (grant No. 1027/18).  ... 
dblp:conf/usenix/BrokhmanLS19 fatcat:imcpoceqtrbgzplty7ul2xltx4

GPUfs

Mark Silberstein, Bryan Ford, Emmett Witchel
2014 Communications of the ACM  
of system abstractions such as data movement and management operations.  ...  Today's GPU is capable of amazing computational feats when fed with the right data and managed by application code on the host CPU, but it is incapable of initiating basic system interactions for itself  ...  Even if the file data is resident in the CPU buffer cache, it must be transferred from CPU memory to the local GPU memory for every program access.  ... 
doi:10.1145/2656206 fatcat:oy4unsirhnbpfbab2pptkz44j4
« Previous Showing results 1 — 15 out of 7,809 results