2,390 Hits in 3.6 sec

Transaction Routing in Real-Time Shared Disks Clusters [chapter]

Kyungoh Ohn, Sangho Lee, Haengrae Cho
2005 Lecture Notes in Computer Science  
This paper proposes a real-time transaction routing algorithm, which allocates real-time transactions to a node in a shared disks cluster.  ...  Unlike traditional routing algorithms, which consider load balancing and transaction affinity only, our algorithm also considers transaction priorities inherent to real-time applications.  ...  Introduction Although there has been a great deal of independent research in real-time processing and shared disks (SD) cluster, their aggregation has very little attention [3] .  ... 
doi:10.1007/11428862_160 fatcat:6jkxhlxnf5dczgmia5p6spgw7q

Cluster architectures and S/390 Parallel Sysplex scalability

G. M. King, D. M. Dias, P. S. Yu
1997 IBM Systems Journal  
It also provides for effective dynamic load balancing, high data buffer hit ratios, and load balancing after a failure.  ...  We outline various cluster architecture% and describe the factors that motivate the SI390 Parallel SysplexTM architecture and its resulting advantages.  ...  The shared-disk architecture is illustrated in Figure 3 . 6,16~17 Essentially, all the nodes in the clustered system have direct access to (some or all) of the disks on which shared data are placed.  ... 
doi:10.1147/sj.362.0221 fatcat:rp3f375iyjhcrnt7gucaowsroq

Analysis of parallel scan processing in Shared Disk database systems [chapter]

Erhard Rahm, Thomas Stöhr
1995 Lecture Notes in Computer Science  
Finally, we show that disk contention in multi-user mode can be limited for Shared Disk database systems by dynamically choosing the degree of scan parallelism.  ...  We present a detailed simulation study to analyze the effectiveness of parallel scan processing in Shared Disk database systems.  ...  Note that such a dynamic query scheduling approach is feasible for Shared Disk, but not for Shared Nothing.  ... 
doi:10.1007/bfb0020487 fatcat:njrgyqxdfzf37jmdyarp26wcdy

Towards Balance-Affinity Tradeoff in Concurrent Subgraph Traversals

Yinglong Xia, Lifeng Nai, Jui-Hsin Lai
2015 2015 IEEE International Parallel and Distributed Processing Symposium  
A dynamic weighted bipartite graph is built to model the affinity between subgraph traversals and processors, and the workload of processors.  ...  In this paper, we present an auction based approach for allocating concurrent subgraph traversals onto the processors.  ...  It may require additional clustering to form an allocation scheme. (3) The concurrent queries may keep on streaming in. The above method is of limited capability to handle the dynamic queries.  ... 
doi:10.1109/ipdps.2015.25 dblp:conf/ipps/XiaNL15 fatcat:ng62xgbuzfg6fode5jnzj5hzti

The VirtuaLinux Storage Abstraction Layer for Ef?cient Virtual Clustering

Marco Aldinucci, Massimo Torquati, Marco Vanneschi, Pierfrancesco Zuccato
2008 16th Euromicro Conference on Parallel, Distributed and Network-Based Processing (PDP 2008)  
VirtuaLinux helps in avoiding the "single point of failure" effect by means of a combination of architectural strategies, including the transparent support for disk-less and master-less cluster configuration  ...  , and to map and dynamically re-map them onto the nodes of the physical cluster.  ...  The configuration is dynamically updated in the case virtual nodes are re-mapped onto the physical cluster.  ... 
doi:10.1109/pdp.2008.86 dblp:conf/pdp/AldinucciTVZ08 fatcat:7ow3n24ebnh6fjz5xqcgivd6zm

Taming Performance Hotspots in Cloud Storage with Dynamic Load Redistribution

Ridwan Rashid Noel, Palden Lama
2017 2017 IEEE 10th International Conference on Cloud Computing (CLOUD)  
In this paper, we present DLR, a system that improves the performance of cloud storage services in the presence of hardware heterogeneity, and performance interference through a dynamic load redistribution  ...  This can be attributed to performance hotspots created by slow nodes in a storage cluster, and performance interference caused by multi-tenancy, and background tasks such as data scrubbing, backfilling  ...  ACKNOWLEDGMENT Results presented in this paper were obtained using the Chameleon testbed supported by the National Science Foundation.  ... 
doi:10.1109/cloud.2017.15 dblp:conf/IEEEcloud/NoelL17 fatcat:xoryw66gcfdyzif37rmuyzomau

MIMP: Deadline and Interference Aware Scheduling of Hadoop Virtual Machines

Wei Zhang, Sundaresan Rajasekaran, Timothy Wood, Mingfa Zhu
2014 2014 14th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing  
prevents this from being a practical solution in virtualized environments.  ...  We have implemented our system using Xen and Hadoop, and our evaluation shows that our schedulers allow a mixed cluster to reduce web response times by more than ten fold, while meeting more Hadoop deadlines  ...  something in between for a shared slot allocated to this job.  ... 
doi:10.1109/ccgrid.2014.101 dblp:conf/ccgrid/ZhangRWZ14 fatcat:jtrr6mzphbgxnckzssfvid7npm

On the core affinity and file upload performance of Hadoop

Joong-Yeon Cho, Hyun-Wook Jin, Min Lee, Karsten Schwan
2013 Proceedings of the 2013 International Workshop on Data-Intensive Scalable Computing Systems - DISCS-2013  
We also suggest a framework for dynamic core affinity based on our observations and show that a preliminary implementation can improve the throughput more than 40% compared with default Linux system.  ...  The core affinity defines mapping between a set of cores and a given task.  ...  DYNAMIC CORE AFFINITY FOR HDFS In this section, we suggest a framework of dynamic core affinity for HDFS. The overall design of the framework is shown in Figure 5 .  ... 
doi:10.1145/2534645.2534651 dblp:conf/sc/ChoJLS13 fatcat:vozfjbkh6zbnvme3rykqjeyemy

Dynamic Allocation of High Performance Computing Resources

Manish Kumar Abhishek
2020 International Journal of Advanced Trends in Computer Science and Engineering  
This paper presents a model which will help in dynamically allocation of the computational resources in terms of CPU, memory to overcome the existing challenges and reduce performance overhead.  ...  Several tools are available for cluster management, but resource usage and consumption are always a concern.  ...  Mesos It is a cluster resource manager which provides isolation and resource sharing among distributed jobs or applications.  ... 
doi:10.30534/ijatcse/2020/159932020 fatcat:2bqmqjpfefbp7cprgkbarlwj4u

On caching effectiveness of web clusters under persistent connections

Xueyan Tang, Samuel T. Chanson
2003 Journal of Parallel and Distributed Computing  
In general, a higher cache hit ratio implies lower disk load and higher system throughput.  ...  Based on a heuristic approach, a session-affinity aware algorithm is presented that makes use of the correlation between the requests in a session.  ...  Acknowledgements The work described in this paper was supported by a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (Project No. HKUST6066/00E).  ... 
doi:10.1016/s0743-7315(03)00098-4 fatcat:l3cx5xpnjfa3zejybijlulfqfa

Advancements in YARN Resource Manager [chapter]

Konstantinos Karanasos, Arun Suresh, Chris Douglas
2018 Encyclopedia of Big Data Technologies  
This allows several applications, including MapReduce, to be deployed on a single cluster and share the same resource management layer.  ...  Synonyms Cluster scheduling; Job scheduling; Resource management Definitions YARN is currently one of the most popular frameworks for scheduling jobs and managing resources in shared clusters.  ...  requirements over large shared clusters.  ... 
doi:10.1007/978-3-319-63962-8_207-1 fatcat:ytb2h4tr2ncjjdv2znh6plfkzm

Starling: Minimizing Communication Overhead in Virtualized Computing Platforms Using Decentralized Affinity-Aware Migration

Jason Sonnek, James Greensky, Robert Reutiman, Abhishek Chandra
2010 2010 39th International Conference on Parallel Processing  
In this paper, we present a decentralized affinity-aware migration technique that incorporates heterogeneity and dynamism in network topology and job communication patterns to allocate virtual machines  ...  Our experimental results running the Intel MPI benchmark and a scientific application on a 7-node Xen cluster show that we can get up to 42% improvement in the runtime of the application over a no-migration  ...  In this paper, we presented a decentralized affinity-aware migration technique that incorporates heterogeneity and dynamism in network topology and job communication patterns to allocate virtual machines  ... 
doi:10.1109/icpp.2010.30 dblp:conf/icpp/SonnekGRC10 fatcat:4rdq56mp2jg65aqrtor4xudbbm

Electronic Structure Calculations and Adaptation Scheme in Multi-core Computing Environments [chapter]

Lakshminarasimhan Seshagiri, Masha Sosonkina, Zhao Zhang
2009 Lecture Notes in Computer Science  
In this paper we try to benchmark the GAMESS performance on a T2 Niagara processor for a couple of molecules.  ...  The T2 Niagara processor is a very unique environment where it consists of eight cores having a capability of running eight threads simultaneously in each of the cores.  ...  These could include strategies such as changing the thread allocation dynamically from a single core to span multiple cores and to take advantage of processor affinity.  ... 
doi:10.1007/978-3-642-01970-8_1 fatcat:s4owh75hybfipge2tduawxjuwa

S/390 cluster technology: Parallel Sysplex

J. M. Nick, B. B. Moore, J.-Y. Chung, N. S. Bowen
1997 IBM Systems Journal  
A DEFINE VECTOR instruction dynamically allocates, deallocates, or changes the size of a local state vector.  ...  Repar- titioning of the cluster databases to introduce new cluster nodes for additional capacity requires the en- tire cluster to be shut down. The shared-disk (shared-data) model.  ... 
doi:10.1147/sj.362.0172 fatcat:ggd2f6u5qvhypgys6lwwflpvhe


Khaled Elmeleegy, Christopher Olston, Benjamin Reed
2014 Proceedings of the 2014 ACM SIGMOD international conference on Management of data - SIGMOD '14  
A SpongeFile is a logical byte array, comprised of large chunks that can be stored in a variety of locations in the cluster.  ...  Moreover, performance of other jobs sharing same disk degrades. In many cases, this situation occurs even as the cluster has plenty of spare memory-it is just not used evenly.  ...  to the dynamic distributed environments of MapReduce clusters.  ... 
doi:10.1145/2588555.2595634 dblp:conf/sigmod/ElmeleegyOR14 fatcat:6vrrwd2kcvcjjdumgpr2nxz7de
« Previous Showing results 1 — 15 out of 2,390 results