74,974 Hits in 5.6 sec

A Comparative Analysis of Performance of Shared Memory Cluster Computing Interconnection Systems

Minakshi Tripathy, C. R. Tripathy
2014 Journal of Computer Networks and Communications  
In this paper, a detailed comparative study on four important and different classes of shared memory cluster architectures has been made.  ...  The systems taken up for the purpose of the study are shared memory clusters, hierarchical shared memory clusters, distributed shared memory clusters, and the virtual distributed shared memory clusters  ...  Conflict of Interests The authors declare that there is no conflict of interests regarding the publication of this paper. Journal of Computer Networks and Communications  ... 
doi:10.1155/2014/128438 fatcat:qcjwpzai3bewnnvkti7s6vu7tq

Retrospective: the Cedar system

A. Veidenbaum, P.-C. Yew, D. J. Kuck, C. D. Polychronopoulos, D. H. Padua, E. S. Davidson, K. Gallivan
1998 25 years of the international symposia on Computer architecture (selected papers) - ISCA '98  
Another goal was to have the prototype "achieve Cray-1 speeds for programs written in high-level languages and automatically restructured" by a compiler.  ...  accepted as a way to speed up the execution of a single program.  ...  File systems in each cluster were made available to the user as a single abstraction.  ... 
doi:10.1145/285930.285965 dblp:conf/isca/VeidenbaumYKPPDG98 fatcat:obv6ipwmffbh7e27pzrhkzaza4

A Hierarchical Shared Memory Cluster Architecture with Load Balancing and Fault Tolerance

Minakshi Tripathy, C.R. Tripathy
2011 International Journal of Computer Applications  
Recently a great deal of attention has been paid to the design of hierarchical shared memory cluster system.  ...  The hierarchical load balancing approach focuses on reducing the redistribution cost. The fault tolerant model is adopted to build highly available clusters in hierarchical shared memory clusters.  ...  hierarchy. iii) Cluster shared memory (Ms 1 ) globally shared by processors in each cluster at the 1st level of hierarchy.  ... 
doi:10.5120/3038-4121 fatcat:ad5ngyzz7be3dmn42t67iqorhi

Low Power and Scalable Many-Core Architecture for Big-Data Stream Computing

Karim Kanoun, Martino Ruggiero, David Atienza, Mihaela van der Schaar
2014 2014 IEEE Computer Society Annual Symposium on VLSI  
Moreover, our platform architecture includes a memory hierarchy optimized for Big-Data streaming and implements modern fine-grained power management techniques over all the different types of cores allowing  ...  In the last years the process of examining large amounts of different types of data, or Big-Data, in an effort to uncover hidden patterns or unknown correlations has become a major need in our society.  ...  Our results for machine learning algorithms have demonstrated that our solution allows to significantly reduce the data memory access conflicts even for a high number of cores in the cluster.  ... 
doi:10.1109/isvlsi.2014.77 dblp:conf/isvlsi/KanounRAS14 fatcat:vu3tpx4wffa6fc5kiteuf6gnxa

What's Up with the Storage Hierarchy?

Philippe Bonnet
2017 Conference on Innovative Data Systems Research  
Today, who can make predictions about the future of the storage hierarchy? Both main memory and storage systems are undergoing profound transformations.  ...  The impact of these transformations on the storage hierarchy is unclear. Yet, they raise interesting research questions.  ...  On a cluster, the storage hierarchy is stretched across two dimensions (memory/disk, local/rack/cluster) and offers a range of possible trade-offs in terms of latency, bandwidth and capacity [3] .  ... 
dblp:conf/cidr/Bonnet17 fatcat:ji4pvfutljcpdfncnexv2gjtum

Bridging the Gap Between Cluster and Grid Computing [chapter]

Albano Alves, António Pina
2006 Lecture Notes in Computer Science  
The Internet computing model with its ubiquitous networking and computing infrastructure is driving a new class of interoperable applications that benefit both from high computing power and multiple Internet  ...  Ro c meµ is a platform originally designed to support the operation of multi-SAN clusters that integrates application modeling and resource allocation.  ...  Next, we describe the mapping process used to produce the hierarchy presented in figure 5 , which corresponds to the fusion of two hierarchies -the hierarchy that represents the cluster ( figure 4(a)  ... 
doi:10.1007/11752578_25 fatcat:z7sixbkmovfg5lpcfdbd6chjui

Design and evaluation of a TOP100 Linux Super Cluster system

Niklas Edmundsson, Erik Elmroth, Bo Kågström, Markus Mårtensson, Mats Nylén, Åke Sandgren, Mattias Wadenstein
2004 Concurrency and Computation  
The High Performance Computing Center North (HPC2N) Super Cluster is a truly self-made high-performance Linux cluster with 240 AMD processors in 120 dual nodes, interconnected with a high-bandwidth, low-latency  ...  In summary, this $500 000 system is extremely cost-effective and shows the performance one would expect of a large-scale supercomputing system with distributed memory architecture.  ...  The procurement of the hardware was made possible by a grant from The Kempe Foundations.  ... 
doi:10.1002/cpe.787 fatcat:h2ulkmmqendhdfdb3f2eam6sry

Deploying applications in multi-SAN SMP clusters

Albano Alves, Antonio Pina, Jose Exposto, Jose Rufino
2009 International Journal of Computational Science and Engineering (IJCSE)  
In this paper we present a novel approach to the exploitation of clusters that allows integrating in a unique metaphor: the representation of physical resources, the modelling of applications and the mapping  ...  The proposed abstractions favoured the development of an API that allows combining and benefiting from the shared memory, message passing and global memory paradigms. .  ...  Global memory A fundamental issue in cluster computing is memory hierarchy and as a consequence the exploitation of data locality is one of the keys to achieve high-performance.  ... 
doi:10.1504/ijcse.2009.027376 fatcat:6ff45oddena2bp3axmenp7en4m

Hierarchical Dataflow Model for efficient programming of clustered manycore processors

Julien Hascoet, Karol Desnos, Jean-Francois Nezan, Benoit Dupont de Dinechin
2017 2017 IEEE 28th International Conference on Application-specific Systems, Architectures and Processors (ASAP)  
Deployment of an image processing application on a many-core MPSoC results in speedups of up to 58.7 compared to the sequential execution.  ...  Dataflow Models of Computation (MoCs) are increasingly used for developing parallel applications as their high-level of abstraction eases the automation of mapping, task scheduling and memory allocation  ...  High-Level Hierarchy (Inter-Cluster) The high-level mapping of hierarchical actors is more efficient as it is adapted for coarse-grained granularity.  ... 
doi:10.1109/asap.2017.7995270 dblp:conf/asap/HascoetDND17 fatcat:zrl6ffctonhrzixhzhh35nbzdy

Modeling and Simulative Performance Analysis of SMP and Clustered Computer Architectures

Mark W. Burns, Alan D. George, Brad A. Wallace
2000 Simulation (San Diego, Calif.)  
Due to the high fidelity of our models, these low-level performance statistics are more practically gathered than in testbed analyses yet more accurate than the results of purely analytical models.  ...  Processor, bus, and network models are used to construct and simulate the architectures of symmetric multiprocessors (SMPs), clusters of uniprocessors, and clusters of SMPs.  ...  A cluster of SMPs attempts to garner the benefits of locality and cost-effectiveness available in SMPs while also benefiting from the removal of the bottleneck and limits on scalability associated with  ... 
doi:10.1177/003754970007400203 fatcat:t3aac62snrchdb4yojb5fnuvja

POSTER: Leveraging deep memory hierarchies for data staging in coupled data-intensive simulation workflows

Tong Jin, Fan Zhang, Qian Sun, Hoang Bui, Norbert Podhorszki, Scott Klasky, Hemanth Kolla, Jacqueline Chen, Robert Hager, Choong-Seock Chang, Manish Parashar
2014 2014 IEEE International Conference on Cluster Computing (CLUSTER)  
However, further research is necessary in order to understand how growing data sizes from data intensive simulations coupled with limited DRAM capacity in High End Computing clusters will impact the effectiveness  ...  In this work, we propose using deep memory levels for data staging, utilizing a multi-tiered data staging method with both DRAM and solid state disk (SSD).  ...  Furthermore, it ensures that data is moved to DRAM prior to a read request in order to avoid the penalty of reading data from a lower level of the memory hierarchy, i.e., the SSD level, and improving the  ... 
doi:10.1109/cluster.2014.6968744 dblp:conf/cluster/JinZSBPKKCHCP14 fatcat:z63di74hzzcllgxn3gy6cwvuv4

Stream Scheduling: A Framework to Manage Bulk Operations in a Memory Hierarchy

Abhishek Das, William J. Dally
2007 Parallel Architecture and Compilation Techniques (PACT), Proceedings of the International Conference on  
The levels of memory hierarchy could include on-die storage such as caches, local DRAM, or even remote memory accessed over high-speed interconnect.  ...  the memory hierarchy.  ...  Stream Scheduling optimizes the use of available memory at each level of the hierarchy by varying a collection of parameters(tunables) in the Sequoia program, used with blocking primitives for task decomposition  ... 
doi:10.1109/pact.2007.4336233 fatcat:m2s7y4urdjc3niwi2g4e6fwtbe


Kayvon Fatahalian, William J. Dally, Pat Hanrahan, Daniel Reiter Horn, Timothy J. Knight, Larkhoon Leem, Mike Houston, Ji Young Park, Mattan Erez, Manman Ren, Alex Aiken
2006 Proceedings of the 2006 ACM/IEEE conference on Supercomputing - SC '06  
We present Sequoia, a programming language designed to facilitate the development of memory hierarchy aware parallel programs that remain portable across modern machines featuring different memory hierarchy  ...  Sequoia abstractly exposes hierarchical memory in the programming model and provides language mechanisms to describe communication vertically through the machine and to localize computation to particular  ...  DRAM, or remote memory accessed over high-speed interconnect.  ... 
doi:10.1145/1188455.1188543 dblp:conf/sc/FatahalianHKLHPERADH06 fatcat:vvqqskhcqbbopiv2q4ha5yvda4

A catalog of classifying characteristics for massively parallel computers [chapter]

Tilmann Bönniger, Rüdiger Esser, Dietrich Krekel
1994 Lecture Notes in Computer Science  
The hardware model The catalog is based on a generic hardware model (Figure 1 ) that most of today's commercial MP systems seem to follow: a three-stage hierarchy of processing nodes, clusters of nodes  ...  In order to facilitate an application-oriented assessment of high-performance massively parallel computing systems, a catalog of about 350 classifying characteristics concerning the architecture and software  ...  memory hierarchy.  ... 
doi:10.1007/3-540-57981-8_102 fatcat:q7twxkwe2jccpbm6tmd2kx5vsa

Exposing the Locality of Heterogeneous Memory Architectures to HPC Applications

Brice Goglin
2016 Proceedings of the Second International Symposium on Memory Systems - MEMSYS '16  
High-performance computing requires a deep knowledge of the hardware platform to fully exploit its computing power. The performance of data transfer between cores and memory is becoming critical.  ...  Its memory locality cannot be properly exposed to user-space applications without a significant rework of the current software stack.  ...  Memory Architecture of the KNL From the memory point of view, the main innovation in the Knights Landing processor is the integration of a high-bandwidth memory in the package [18] .  ... 
doi:10.1145/2989081.2989115 dblp:conf/memsys/Goglin16 fatcat:eev2v2bomzcdri2gnsnfyn3fey
« Previous Showing results 1 — 15 out of 74,974 results