LogP

Culler, David; Karp, Richard M.; Patterson, David A.; Sahay, Abhijit; Santos, Eunice E.; Schauser, Klaus E.; Subramonian, Ramesh; Eicken, Thorsten von

doi:10.1145/240455.240477

Cited by 245 publications

(108 citation statements)

References 22 publications

Supporting

Mentioning

103

Contrasting

Unclassified

Order By: Relevance

“…Still, innermost recursions need to be unfolded [2], just as the C compiler unrolls loops, in order to obtain straight-line code for superscalar processors. See [18] for more details; all the speed improvements are due to good locality [5].…”

Section: Resultsmentioning

confidence: 99%

Ahnentafel Indexing into Morton-Ordered Arrays, or Matrix Locality for Free

Wise

2000

Euro-Par 2000 Parallel Processing

View full text Add to dashboard Cite

Abstract. Definitions for the uniform representation of d-dimensional matrices serially in Morton-order (or Z-order) support both their use with cartesian indices, and their divide-and-conquer manipulation as quaternary trees. In the latter case, d-dimensional arrays are accessed as 2 d -ary trees. This data structure is important because, at once, it relaxes serious problems of locality and latency, and the tree helps schedule multiprocessing. It enables algorithms that avoid cache misses and page faults at all levels in hierarchical memory, independently of a specific runtime environment.This paper gathers the properties of Morton order and its mappings to other indexings, and outlines for compiler support of it. Statistics elsewhere show that the new ordering and block algorithms achieve high flop rates and, indirectly, parallelism without any low-level tuning.

show abstract

Section: Resultsmentioning

confidence: 99%

Ahnentafel Indexing into Morton-Ordered Arrays, or Matrix Locality for Free

Wise

2000

Euro-Par 2000 Parallel Processing

View full text Add to dashboard Cite

show abstract

“…Generating queries for resource discovery requires a capability to predict the resource requirements of parallel applications before making decisions for scheduling. There are a multitude of technologies to estimate and extract the application features which try to deploy performance models or mechanisms to forestall resource utilization in case of computation and communication complexities for applications launched under a distributed parallel system including multiple homogeneous or heterogeneous resources with various processing capabilities [12,13]. Application modeling or application description can provide accurate information for querying operations in the resource discovery protocol.…”

Section: Multi-dimensional Discoverymentioning

confidence: 99%

Dynamic, scalable and flexible resource discovery for large-dimension many-core systems

Zarrin

Aguiar

Barraca

2015

Future Generation Computer Systems

View full text Add to dashboard Cite

This is the accepted version of the paper.This version of the publication may differ from the final published version. Abstr act Future large scale systems will execute novel operating systems running across many chips with many cores. In this highly distributed environment, resource discovery is an important building block. Resource discovery aims to match the application's demands to the existing (distributed) resources, by discovering and f nding resources at run-time, and then selecting the best resource that matches the application running requirements. The main contribution of this paper is the design and evolution of a highly scalable, highly f exible, resource discovery model for such heterogeneous environments. The model is based on self-organizing processing resources in the system according to a hierarchical resource description where each group of resources has a local directory that collects and keeps the information of the underlying resource members (cores) in dif erent layers. Operationally, at each layer, it consists of a peer-to-peer architecture of modules that, by interacting with each other, provide a global view of the resource availability in a large, dynamic and heterogeneous distributed environment. The proposed resource discovery model provides the adaptability and f exibility to perform complex querying by supporting a large set of signif cant querying features (such as multi-dimensional, range and aggregate querying) while supporting exact and partial matching, both for static and dynamic object contents. The paper demonstrates by simulation how the proposed model can deal with issues such as scalability, ef ciency and adaptability of resource discovery in future many-core systems which are the major challenges in the current state of the art. The simulation show that the proposed resource discovery model can be applied to arbitrary scales of dynamicity, both in terms of complexity and of scale, positioning this proposal as a good architecture for future many-core systems. Permanent repository link

show abstract

“…Due to the emergence of network of workstations as high performance environment, the LogP model [24] was proposed to be a computational model in which global characteristics of parallel architectures are represented, such as number of processing elements, latency on the transmissions, gap between subsequent messages and overhead on the sending and receiving of messages. The key issues stated in the model were related to communication and non-synchronous computations.…”

Section: Model For Distributed Memory Architecturementioning

confidence: 99%

Memory aware load balance strategy on a parallel branch‐and‐bound application

Silva

Boeres

Drummond

et al. 2014

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARYThe latest trends in high-performance computing systems show an increasing demand on the use of a large scale multicore systems in a efficient way, so that high compute-intensive applications can be executed reasonably well. However, the exploitation of the degree of parallelism available at each multicore component can be limited by the poor utilization of the memory hierarchy available. Actually, the multicore architecture introduces some distinct features that are already observed in shared memory and distributed environments. One example is that subsets of cores can share different subsets of memory. In order to achieve high performance it is imperative that a careful allocation scheme of an application is carried out on the available cores, based on a scheduling model that considers the main performance bottlenecks, as for example, memory contention. In this paper, the Multicore Cluster Model (MCM) is proposed, which captures the most relevant performance characteristics in multicores systems such as the influence of memory hierarchy and contention. Better performance was achieved when a load balance strategy for a Branch-and-Bound application applied to the Partitioning Sets Problem is based on MCM, showing its efficiency and applicability to modern systems.

show abstract

LogP

Cited by 245 publications

References 22 publications

Ahnentafel Indexing into Morton-Ordered Arrays, or Matrix Locality for Free

Ahnentafel Indexing into Morton-Ordered Arrays, or Matrix Locality for Free

Dynamic, scalable and flexible resource discovery for large-dimension many-core systems

Memory aware load balance strategy on a parallel branch‐and‐bound application

Contact Info

Product

Resources

About