5,631 Hits in 6.1 sec

Performance Sensitive Replication in Geo-distributed Cloud Datastores

P.N. Shankaranarayanan, Ashiwan Sivakumar, Sanjay Rao, Mohit Tawarmalani
2014 2014 44th Annual IEEE/IFIP International Conference on Dependable Systems and Networks  
We evaluate our models using real-world traces of three applications: Twitter, Wikipedia and Gowalla on a Cassandra cluster deployed in Amazon EC2.  ...  We present models that optimize percentiles of response time under normal operation and under a data-center (DC) failure.  ...  A key observation is that, given the replica locations, all read and, similarly all write requests, that originate from a given DC encounter the same delay.  ... 
doi:10.1109/dsn.2014.34 dblp:conf/dsn/NSRT14 fatcat:dmi3mbl7zba2dkjwet3j46vcd4


Elena Kakoulli, Herodotos Herodotou
2017 Proceedings of the 2017 ACM International Conference on Management of Data - SIGMOD '17  
While past work explores using memory or SSDs as local storage or combine local with network-attached storage in cluster computing, this work focuses on managing multiple storage tiers in a distributed  ...  The policies employ multi-objective optimization techniques for making intelligent data management decisions based on the requirements of fault tolerance, data and load balancing, and throughput maximization  ...  The order is determined using the network location of the Client, the network topology of the Workers, as well as the storage tiers that host the replicas.  ... 
doi:10.1145/3035918.3064023 dblp:conf/sigmod/KakoulliH17 fatcat:lneuhofn4fdgdn4t5e6int2f5y

Locality-aware data replication in the Last-Level Cache

George Kurian, Srinivas Devadas, Omer Khan
2014 2014 IEEE 20th International Symposium on High Performance Computer Architecture (HPCA)  
Harnessing on-chip data locality to optimize the utilization of cache and network resources is of fundamental importance.  ...  Furthermore, our classifier captures the LLC pressure at the existing replica locations and adapts its replication decision accordingly.  ...  Hence, in the common case where replication is useful, each hit at the LLC location effectively incurs both a read and a write at the LLC. And a write expends 1.2× more energy than a read.  ... 
doi:10.1109/hpca.2014.6835921 dblp:conf/hpca/KurianDK14 fatcat:gsjpmhtpvzbpdjv6fdmvwfqt6m

An Efficient Group-Based Replica Placement Policy for Large-Scale Geospatial 3D Raster Data on Hadoop

Zhipeng Liu, Weihua Hua, Xiuguo Liu, Dong Liang, Yabo Zhao, Manxing Shi
2021 Sensors  
To address this problem, we propose an efficient group-based replica placement policy for large-scale geospatial 3D raster data, aiming to optimize the locations of the replicas in the cluster to reduce  ...  As a result, the direct processing of geospatial 3D raster data produces a massive network data exchange among the datanodes and degrades the performance of the cluster.  ...  Conflicts of Interest: The authors declare no conflict of interest. Sensors 2021, 21, 8132  ... 
doi:10.3390/s21238132 pmid:34884135 fatcat:zl4sekam65am3celpxrlscuc74

DynaSoRe: Efficient In-Memory Store for Social Applications [chapter]

Xiao Bai, Arnaud Jégou, Flavio Junqueira, Vincent Leroy
2013 Lecture Notes in Computer Science  
In this work, we present DynaSoRe: a memory cache system for social network applications that optimizes data locality while placing user views across the system.  ...  To enable fast responses to user requests, social network applications typically rely on large banks of cache servers to hold and serve most of their content from the cache.  ...  This work was supported by the LEADS project (ICT-318809), funded by the European Community, the Torres Quevedo Program from the Spanish Ministry of Science and Innovation, co-funded by the European Social  ... 
doi:10.1007/978-3-642-45065-5_22 fatcat:2g3kvhhn5bhl7hsk623mdf7aye

Take me to your leader!

Artyom Sharov, Alexander Shraer, Arif Merchant, Murray Stokely
2015 Proceedings of the VLDB Endowment  
) roles of different servers in the replication protocol, and 3) replica locations.  ...  The configuration of a distributed storage system typically includes, among other parameters, the set of servers and their roles in the replication protocol.  ...  Since all groups in a database are replicated in the same way, this method reduces the solution space to one of the read-write replica locations for this database.  ... 
doi:10.14778/2824032.2824047 fatcat:av6ri36rczg6zjabmdnmwslifa

Leveraging endpoint flexibility in data-intensive clusters

Mosharaf Chowdhury, Srikanth Kandula, Ion Stoica
2013 Proceedings of the ACM SIGCOMM 2013 conference on SIGCOMM - SIGCOMM '13  
The replicas of a CFS write can be placed in any subset of machines as long as they are in multiple fault domains and ensure a balanced use of storage throughout the cluster.  ...  In this paper, we focus on leveraging the flexibility in replica placement during writes to cluster file systems (CFSes), which account for almost half of all cross-rack traffic in data-intensive clusters  ...  useful feedback.  ... 
doi:10.1145/2486001.2486021 dblp:conf/sigcomm/ChowdhuryKS13 fatcat:kz2tsvemifdc5eawzcsyhaywe4

Leveraging endpoint flexibility in data-intensive clusters

Mosharaf Chowdhury, Srikanth Kandula, Ion Stoica
2013 Computer communication review  
The replicas of a CFS write can be placed in any subset of machines as long as they are in multiple fault domains and ensure a balanced use of storage throughout the cluster.  ...  In this paper, we focus on leveraging the flexibility in replica placement during writes to cluster file systems (CFSes), which account for almost half of all cross-rack traffic in data-intensive clusters  ...  useful feedback.  ... 
doi:10.1145/2534169.2486021 fatcat:554v7bptlbc77erg4jckkx2xum

Placement of File Replicas in Data Grid Environments [chapter]

J. H. Abawajy
2004 Lecture Notes in Computer Science  
In this paper, we address the problem of file replica placement in Data Grids given a certain traffic pattern.  ...  The results show that file replication improve the performance of the data access but the gains depend on several factors including where the file replicas are located, burstness of the request arrival  ...  Therefore, given the network topology and data request patterns shown in Figure 1 , the goal is to determine an optimal placement of a set of f file replicas on the n sites such that data access delays  ... 
doi:10.1007/978-3-540-24688-6_11 fatcat:k5bwvowzrrcvfkyqrqfx5himva

OctopusFS in action

Elena Kakoulli, Nikolaos D. Karmiris, Herodotos Herodotou
2018 Proceedings of the VLDB Endowment  
It also exposes the network locations and storage tiers of the data in order to allow higher-level systems to make locality-aware and tier-aware decisions.  ...  The continuous improvements in memory, storage devices, and network technologies of commodity hardware introduce new challenges and opportunities in tiered storage management.  ...  The default one (presented in [4] ) takes into consideration the Client location, the replica locations, the network topology, the average data transfer rates from each Worker, the read throughput rate  ... 
doi:10.14778/3229863.3236223 fatcat:ioql5kdbujea7icgruj5v6dcoi

Scalable strong consistency for web applications

Swaminathan Sivasubramanian, Guillaume Pierre, Maarten van Steen
2004 Proceedings of the 11th workshop on ACM SIGOPS European workshop: beyond the PC - EW11  
Web application workloads are often characterized by a large number of unique read requests and a significant fraction of write requests.  ...  This reduces the consistency overhead as updates are sent to a reduced number of servers.  ...  This limits the system performance as all write requests need to be forwarded to a single remote location.  ... 
doi:10.1145/1133572.1133581 dblp:conf/sigopsE/SivasubramanianPS04 fatcat:ce2vibb52balvbfd7cc3zl4gla

Stable Neighbor Based Adaptive Replica Allocation in Mobile Ad Hoc Networks [chapter]

Zheng Jing, Su Jinshu, Yang Kan, Wang Yijie
2004 Lecture Notes in Computer Science  
This algorithm dynamically adjusts the replica allocation scheme towards a local optimal one.  ...  In mobile ad hoc networks (MANET), nodes move freely and the replica allocation in such a dynamic environment is a significant challenge.  ...  The ROWA (READ-ONE-WRITE-ALL) policy is used to ensure the consistency of the replicas, and we assume that each individual access is independent.  ... 
doi:10.1007/978-3-540-24685-5_47 fatcat:nog5gg467ncpzct44r4xrxzwja

Placement problems for transparent data replication proxy services

Jianliang Xu, Bo Li, Dik Lun Lee
2002 IEEE Journal on Selected Areas in Communications  
To address the placement problems, we first present the optimal solutions for a single object in a tree network without/with constraint on the number of replicas.  ...  We address the problems of replication proxy placement in the network and data replica placement on the installed proxies given that a maximum of proxies are allowed.  ...  ACKNOWLEDGMENT The authors would like to thank X. Tang for his valuable discussions and the anonymous reviewers for their constructive comments and suggestions.  ... 
doi:10.1109/jsac.2002.802068 fatcat:2v3ula6x75dw7p5ebv2rqiarju

File Clustering Based Replication Algorithm in a Grid Environment

Hitoshi Sato, Satoshi Matsuoka, Toshio Endo
2009 2009 9th IEEE/ACM International Symposium on Cluster Computing and the Grid  
Our experiments on a given grid environment, 20 nodes of 5 sites, suggest that the proposed algorithm achieves accurate file clustering and efficient replica management; our clustering policy with the  ...  Our algorithm groups files according to a relationship of simultaneous accesses between files and stores the replicas of the clustered files into storage nodes, to satisfy expected most of future read  ...  Acknowledgements This research is supported in part by the MEXT Grantin-Aid for Scientific Research on Priority Areas 18049028 and the JSPS Global COE program entitled "Computationism as Foundation for  ... 
doi:10.1109/ccgrid.2009.73 dblp:conf/ccgrid/SatoME09 fatcat:s7vilr2h6fgwda3aox3kvokzam

Database Placement On Large-Scale Systems

Cherif Haddad, Faouzi Ben Charrada
2007 Zenodo  
We evaluate the performance of the proposed database placement policy using simulations.  ...  In this paper, we propose a dynamic database placement policy that depends on query patterns and Grid sites capabilities.  ...  Optimization should be carried out based on the status of Grid resources (workload of computing elements, location of data, network load).  ... 
doi:10.5281/zenodo.1062219 fatcat:paz6arsp3faovau57njosipela
« Previous Showing results 1 — 15 out of 5,631 results