Filters








18,609 Hits in 6.5 sec

TxCache: Transactional cache using byte-addressable non-volatile memories in SSDs

Youyou Lu, Jiwu Shu, Peng Zhu
2014 2014 IEEE Non-Volatile Memory Systems and Applications Symposium (NVMSA)  
In this paper, we propose a new embedded transaction mechanism, TxCache, for SSDs with non-volatile disk cache.  ...  First, it persists new-version data in non-volatile disk cache in a shadow way while protecting old-version data from being overwritten.  ...  TxCache exploits the non-volatility of disk cache for better transaction performance, and clusters pages of each transaction using pointers or flags to minimize transaction overhead.  ... 
doi:10.1109/nvmsa.2014.6927193 dblp:conf/nvmsa/LuSZ14 fatcat:ghh5okkssvctre6ocmhvsc2abi

Transaction Processing Techniques for Modern Hardware and the Cloud

Justin J. Levandoski, Sudipta Sengupta, Ryan Stutsman, Rui Wang
2015 IEEE Data Engineering Bulletin  
data storage and management duties (access methods, cache, stability).  ...  For the past couple of years, we have been rethinking the implementation of both the TC and DC in order for them to perform as well, or better than, current high-performance database systems.  ...  We would like to also thank the various product groups at Microsoft (SQL Server, Azure DocumentDB, and Bing) for helping to improve the Deuteronomy stack.  ... 
dblp:journals/debu/LevandoskiSS015 fatcat:py7wsnfgprfc3f5zpiyswx73xa

Brief Announcement

Mohamed Mohamedin, Roberto Palmieri, Ahmed Hassan, Binoy Ravindran
2015 Proceedings of the 27th ACM on Symposium on Parallelism in Algorithms and Architectures - SPAA '15  
The basic idea of Part-HTM is to partition those transactions into multiple sub-transactions, which can likely be committed in hardware.  ...  In this paper we present Part-HTM, the first hybrid transactional memory protocol that solves the problem of transactions aborted due to the resource limitations (space/time) of current best-effort HTM  ...  ACKNOWLEDGMENTS Authors would like to thank Alexander Matveev for the important feedback. This work is supported in part by US National Science Foundation under grant CNS 1217385.  ... 
doi:10.1145/2755573.2755611 dblp:conf/spaa/MohamedinPHR15 fatcat:o7lhgzqynnet7hmkc7ffbgp3j4

A Multi-version Cache Replacement and Prefetching Policy for Hybrid Data Delivery Environments [chapter]

André Seifert, Marc H. Scholl
2002 VLDB '02: Proceedings of the 28th International Conference on Very Large Databases  
This paper introduces MICP, a novel multiversion integrated cache replacement and prefetching algorithm designed for efficient cache and transaction management in hybrid data delivery networks.  ...  into two variable-sized partitions, namely the REC and the NON-REC partitions for maintaining re-cacheable and nonre-cacheable object versions, respectively.  ...  Now, suppose the client needs to read a transaction-consistent version of object C. Since there is no cache-resident version of object C, the client fetches the missing object from the server.  ... 
doi:10.1016/b978-155860869-6/50080-9 dblp:conf/vldb/SeifertS02 fatcat:pzsub3l3xrejvoxochtommheai

Enhancing recovery using an SSD buffer pool extension

Bishwaranjan Bhattacharjee, Kenneth A. Ross, Christian Lang, George A. Mihaila, Mohammad Banikazemi
2011 Proceedings of the Seventh International Workshop on Data Management on New Hardware - DaMoN '11  
Current work in exploiting flash in database systems has primarily focused on using its random IO capability for second level bufferpools below main memory.  ...  In this paper, we describe a mechanism extending our previous work on a SSD Bufferpool on a DB2 LUW prototype, to exploit the SSD persistence for recovery and normal restart.  ...  In recent years, flash has been exploited as a storage medium for a second-level cache in a DBMS below main memory [3] [7] [13] . For example, multi-level caching using flash is discussed in [5] .  ... 
doi:10.1145/1995441.1995443 dblp:conf/damon/BhattacharjeeRLMB11 fatcat:3xlerzoyvbdohfw2tvqelqallq

Multi-version Indexing in Flash-based Key-Value Stores [article]

Pulkit A. Misra and Jeffrey S. Chase and Johannes Gehrke and Alvin R. Lebeck
2019 arXiv   pre-print
We present SkimpyFTL, a FTL-integrated multi-version key-value store that exploits the remap-on-write property of flash-based SSDs for multi-versioning and provides a tradeoff between memory capacity and  ...  However, designing a multi-version key-value store entails several challenges, such as additional capacity for storing extra versions and an indexing mechanism for mapping versions of a key to their values  ...  Introduction Transactional key-value stores use a multi-version storage to increase concurrency [7] and reduce abort rate [3] as reads can be satisfied from a consistent snapshot in the past (old versions  ... 
arXiv:1912.00580v1 fatcat:vrgw22jjrre7vkwfspbuh7i3oq

Temporally silent stores

Kevin M. Lepak, Mikko H. Lipasti
2002 Tenth international conference on architectural support for programming languages and operating systems on Proceedings of the 10th international conference on architectural support for programming languages and operating systems (ASPLOS-X) - ASPLOS '02  
We redefine multiprocessor sharing to account for temporal silence and show that in the limit, up to 45% of communication misses in scientific and commercial applications can be eliminated by exploiting  ...  Recent work has shown that silent stores--stores which write a value matching the one already stored at the memory location--occur quite frequently and can be exploited to reduce memory traffic and improve  ...  contributions to earlier versions of this work.  ... 
doi:10.1145/605397.605401 dblp:conf/asplos/LepakL02 fatcat:sowmfg7nwze2nh62ahjlpe273m

Temporally silent stores

Kevin M. Lepak, Mikko H. Lipasti
2002 SIGARCH Computer Architecture News  
We redefine multiprocessor sharing to account for temporal silence and show that in the limit, up to 45% of communication misses in scientific and commercial applications can be eliminated by exploiting  ...  Recent work has shown that silent stores--stores which write a value matching the one already stored at the memory location--occur quite frequently and can be exploited to reduce memory traffic and improve  ...  contributions to earlier versions of this work.  ... 
doi:10.1145/635506.605401 fatcat:i4bjcgyvzjeh7l4xcxzhggnide

Temporally silent stores

Kevin M. Lepak, Mikko H. Lipasti
2002 SIGPLAN notices  
We redefine multiprocessor sharing to account for temporal silence and show that in the limit, up to 45% of communication misses in scientific and commercial applications can be eliminated by exploiting  ...  Recent work has shown that silent stores--stores which write a value matching the one already stored at the memory location--occur quite frequently and can be exploited to reduce memory traffic and improve  ...  contributions to earlier versions of this work.  ... 
doi:10.1145/605432.605401 fatcat:ghkmtqxhtjgvverpi5x5buifiq

Temporally silent stores

Kevin M. Lepak, Mikko H. Lipasti
2002 ACM SIGOPS Operating Systems Review  
We redefine multiprocessor sharing to account for temporal silence and show that in the limit, up to 45% of communication misses in scientific and commercial applications can be eliminated by exploiting  ...  Recent work has shown that silent stores--stores which write a value matching the one already stored at the memory location--occur quite frequently and can be exploited to reduce memory traffic and improve  ...  contributions to earlier versions of this work.  ... 
doi:10.1145/635508.605401 fatcat:juimv5gh45ailobeaf2wajjb2q

Temporally silent stores

Kevin M. Lepak, Mikko H. Lipasti
2002 Tenth international conference on architectural support for programming languages and operating systems on Proceedings of the 10th international conference on architectural support for programming languages and operating systems (ASPLOS-X) - ASPLOS '02  
We redefine multiprocessor sharing to account for temporal silence and show that in the limit, up to 45% of communication misses in scientific and commercial applications can be eliminated by exploiting  ...  Recent work has shown that silent stores--stores which write a value matching the one already stored at the memory location--occur quite frequently and can be exploited to reduce memory traffic and improve  ...  contributions to earlier versions of this work.  ... 
doi:10.1145/605398.605401 fatcat:efqkivbrknd5vcrbg4ncbusa5e

High Performance Transactions in Deuteronomy

Justin J. Levandoski, David B. Lomet, Sudipta Sengupta, Ryan Stutsman, Rui Wang
2015 Conference on Innovative Data Systems Research  
(6 million total operations per second) for a variety of workloads.  ...  Importantly, the resulting full stack is a system that caches data residing on secondary storage while exhibiting performance on par with main memory systems.  ...  Version management that exploits the recovery log buffers as a log structured store cache at the TC (Section 3). 5.  ... 
dblp:conf/cidr/LevandoskiLSSW15 fatcat:hjmqm6b7nvh4hmntmll3ey6tu4

TM2C

Vincent Gramoli, Rachid Guerraoui, Vasileios Trigonakis
2012 Proceedings of the 7th ACM european conference on Computer Systems - EuroSys '12  
However, chip manufacturers have started producing many-core architectures, with low network-on-chip communication latency and limited support for cache-coherence, rendering existing transactional memory  ...  Transactional memory is an appealing paradigm for concurrent programming.  ...  Acknowledgments We wish to thank our shepherd, Maurice Herlihy, and the anonymous reviewers for their fruitful comments, and the Intel R MARC Community for its support while programming on the SCC.  ... 
doi:10.1145/2168836.2168872 dblp:conf/eurosys/GramoliGT12 fatcat:z72pzvwq2jbkdhrjc53w6njnuu

Cache Fusion: Extending Shared-Disk Clusters with Shared Caches

Tirthankar Lahiri, Vinay Srihari, Wilson Chan, N. MacNaughton, Sashikanth Chandrasekaran
2001 Very Large Data Bases Conference  
Using Cache Fusion, data blocks are shipped directly from one node to another using interconnect messaging, eliminating the need for extra disk I/Os to facilitate data sharing.  ...  In classic shared-disk implementations, the disk is the medium for data sharing and data blocks are shipped between nodes through disk writes and reads under the arbitration of a distributed lock manager  ...  Cache Fusion Read-Sharing The mechanism for read-sharing in Cache Fusion exploits Oracle's Consistent Read (CR) mechanism [1] .  ... 
dblp:conf/vldb/LahiriSCMC01 fatcat:43xg4v43gzhfpbfp4c4boqdywa

Improving the Performance and Endurance of Persistent Memory with Loose-Ordering Consistency

Youyou Lu, Jiwu Shu, Long Sun, Onur Mutlu
2018 IEEE Transactions on Parallel and Distributed Systems  
To enable this, our mechanism supports the tracking of committed transaction ID and multi-versioning in the CPU cache.  ...  LOC consists of two key techniques. First, Eager Commit eliminates the need to perform a persistent commit record write within a transaction.  ...  To ensure consistency, the program queries the hardware for the persistence status of transactions. BPFS [20] and CLC [83] use versions of this approach.  ... 
doi:10.1109/tpds.2017.2701364 fatcat:6p2yegmf5bamhjhzzhhsbjfd3m
« Previous Showing results 1 — 15 out of 18,609 results