Cache-Oblivious Buffer Heap and Cache-Efficient Computation of Shortest Paths in Graphs

2018 ◽  
Vol 14 (1) ◽  
pp. 1-33 ◽  
Author(s):  
Rezaul A. Chowdhury ◽  
Vijaya Ramachandran
Keyword(s):  
Shortest Paths ◽  
Efficient Computation ◽  
Cache Efficient ◽  
Cache Oblivious

An improved watershed algorithm based on efficient computation of shortest paths

2007 ◽  
Vol 40 (3) ◽  
pp. 1078-1090 ◽  
Author(s):  
Víctor Osma-Ruiz ◽  
Juan I. Godino-Llorente ◽  
Nicolás Sáenz-Lechón ◽  
Pedro Gómez-Vilda
Keyword(s):  
Shortest Paths ◽  
Watershed Algorithm ◽  
Efficient Computation

scholarly journals Efficient Computation of Shortest Paths in Networks Using Particle Swarm Optimization and Noising Metaheuristics

2007 ◽  
Vol 2007 ◽  
pp. 1-25 ◽  
Author(s):  
Ammar W. Mohemmed ◽  
Nirod Chandra Sahoo
Keyword(s):  
Particle Swarm Optimization ◽  
Local Search ◽  
Shortest Path ◽  
Hybrid Algorithm ◽  
Shortest Paths ◽  
Particle Swarm ◽  
Elementary Transformation ◽  
Efficient Computation ◽  
Hybrid Search ◽  
Swarm Optimization

This paper presents a novel hybrid algorithm based on particle swarm optimization (PSO) and noising metaheuristics for solving the single-source shortest-path problem (SPP) commonly encountered in graph theory. This hybrid search process combines PSO for iteratively finding a population of better solutions and noising method for diversifying the search scheme to solve this problem. A new encoding/decoding scheme based on heuristics has been devised for representing the SPP parameters as a particle in PSO. Noising-method-based metaheuristics (noisy local search) have been incorporated in order to enhance the overall search efficiency. In particular, an iteration of the proposed hybrid algorithm consists of a standard PSO iteration and few trials of noising scheme applied to each better/improved particle for local search, where the neighborhood of each such particle is noisily explored with an elementary transformation of the particle so as to escape possible local minima and to diversify the search. Simulation results on several networks with random topologies are used to illustrate the efficiency of the proposed hybrid algorithm for shortest-path computation. The proposed algorithm can be used as a platform for solving other NP-hard SPPs.

scholarly journals Studies on the energy and deep memory behaviour of a cache-oblivious, task-based hyperbolic PDE solver

2019 ◽  
Vol 33 (5) ◽  
pp. 973-986 ◽  
Author(s):  
Dominic E Charrier ◽  
Benjamin Hazelwood ◽  
Ekaterina Tutlyaeva ◽  
Michael Bader ◽  
Michael Dumbser ◽  
...  
Keyword(s):  
Adaptive Mesh Refinement ◽  
Adaptive Mesh ◽  
Main Memory ◽  
Task Graph ◽  
Hyperbolic Pde ◽  
Energy Needs ◽  
Cache Efficient ◽  
Task Types ◽  
Cache Oblivious ◽  
Memory Characteristics

We study the performance behaviour of a seismic simulation using the ExaHyPE engine with a specific focus on memory characteristics and energy needs. ExaHyPE combines dynamically adaptive mesh refinement (AMR) with ADER-DG. It is parallelized using tasks, and it is cache efficient. AMR plus ADER-DG yields a task graph which is highly dynamic in nature and comprises both arithmetically expensive tasks and tasks which challenge the memory’s latency. The expensive tasks and thus the whole code benefit from AVX vectorization, although we suffer from memory access bursts. A frequency reduction of the chip improves the code’s energy-to-solution. Yet, it does not mitigate burst effects. The bursts’ latency penalty becomes worse once we add Intel Optane technology, increase the core count significantly or make individual, computationally heavy tasks fall out of close caches. Thread overbooking to hide away these latency penalties becomes contra-productive with noninclusive caches as it destroys the cache and vectorization character. In cases where memory-intense and computationally expensive tasks overlap, ExaHyPE’s cache-oblivious implementation nevertheless can exploit deep, noninclusive, heterogeneous memory effectively, as main memory misses arise infrequently and slow down only few cores. We thus propose that upcoming supercomputing simulation codes with dynamic, inhomogeneous task graphs are actively supported by thread runtimes in intermixing tasks of different compute character, and we propose that future hardware actively allows codes to downclock the cores running particular task types.

scholarly journals Efficient Computation of Shortest Paths in Time-Dependent Multi-Modal Networks

2015 ◽  
Vol 19 ◽  
pp. 1-29 ◽  
Author(s):  
Dominik Kirchler ◽  
Leo Liberti ◽  
Roberto Wolfler Calvo
Keyword(s):  
Shortest Paths ◽  
Time Dependent ◽  
Efficient Computation

scholarly journals Cache efficient simple dynamic programming

2005 ◽  
Vol DMTCS Proceedings vol. AD,... (Proceedings) ◽  
Author(s):  
Cary Cherng ◽  
Richard E. Ladner
Keyword(s):  
Dynamic Programming ◽  
Dynamic Programming Algorithm ◽  
Recognition Algorithm ◽  
Programming Algorithm ◽  
Language Recognition ◽  
International Audience ◽  
Cache Efficient ◽  
Programming Algorithms ◽  
Context Free ◽  
Cache Oblivious

International audience New cache-oblivious and cache-aware algorithms for simple dynamic programming based on Valiant's context-free language recognition algorithm are designed, implemented, analyzed, and empirically evaluated with timing studies and cache simulations. The studies show that for large inputs the cache-oblivious and cache-aware dynamic programming algorithms are significantly faster than the standard dynamic programming algorithm.

Sign in / Sign up

Export Citation Format

Share Document