Autotuning of Adaptive Mesh Refinement PDE Solvers on Shared Memory Architectures

Nogina, Svetlana; Unterweger, Kristof; Weinzierl, Tobias

doi:10.1007/978-3-642-31464-3_68

Cited by 4 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There is a central instance, a singleton [4] which is notified by the overall algorithm regarding which algorithmic phase is to be run next. We call this instance Oracle [6].…”

Section: Programming An Autotuning Algorithmmentioning

confidence: 99%

“…Our algorithm realises interval halving similar to [6]: We start with relatively large g and compare the multithreaded performance to a serial setting. If the serial version is faster, we deactivate the parallelisation, i.e.…”

Section: Algorithmic Ideamentioning

confidence: 99%

“…Thus, modern parallelisation paradigms allow users to prescribe a grain size, a minimal subproblem size for parallel loops, while taskbased approaches group logical tasks into one physical task if separate tasks were too lightweight. For plain bulk synchronous processing and non-nested tasks, finding grain sizes is often done manually via trial-and-error since developers assume that the size vs. performance curve is convex ([8, p. 37] or [6]).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

An Experience Report on (Auto-)tuning of Mesh-Based PDE Solvers on Shared Memory Systems

Charrier

Weinzierl

2018

Parallel Processing and Applied Mathematics

Self Cite

View full text Add to dashboard Cite

With the advent of manycore systems, shared memory parallelisation has gained importance in high performance computing. Once a code is decomposed into tasks or parallel regions, it becomes crucial to identify reasonable grain sizes, i.e. minimum problem sizes per task that make the algorithm expose a high concurrency at low overhead. Many papers do not detail what reasonable task sizes are, and consider their findings craftsmanship not worth discussion. We have implemented an autotuning algorithm, a machine learning approach, for a project developing a hyperbolic equation system solver. Autotuning here is important as the grid and task workload are multifaceted and change frequently during runtime. In this paper, we summarise our lessons learned. We infer tweaks and idioms for general autotuning algorithms and we clarify that such a approach does not free users completely from grain size awareness.

show abstract

“…There is a central instance, a singleton [4] which is notified by the overall algorithm regarding which algorithmic phase is to be run next. We call this instance Oracle [6].…”

Section: Programming An Autotuning Algorithmmentioning

confidence: 99%

Section: Algorithmic Ideamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Experience Report on (Auto-)tuning of Mesh-Based PDE Solvers on Shared Memory Systems

Charrier

Weinzierl

2018

Parallel Processing and Applied Mathematics

Self Cite

View full text Add to dashboard Cite

show abstract

“…Parallelization within each patch by executing operations on the patch concurrently [31] leads to low scalability for small patch sizes.…”

Section: Simulations With Dynamic Adaptive Mesh Refinementmentioning

confidence: 99%

Invasive Compute Balancing for Applications with Shared and Hybrid Parallelization

Schreiber

Riesinger

Neckel

et al. 2014

Int J Parallel Prog

View full text Add to dashboard Cite

Achieving high scalability with dynamically adaptive algorithms in high-performance computing (HPC) is a non-trivial task. The invasive paradigm using compute migration represents an efficient alternative to classical data migration approaches for such algorithms in HPC. We present a coredistribution scheduler which realizes the migration of computational power by distributing the cores depending on the requirements specified by one or more parallel program instances. We validate our approach with different benchmark suites for simulations with artificial workload as well as applications based on dynamically adaptive shallow water simulations, and investigate concurrently executed adaptivity parameter studies on realistic Tsunami simulations. The invasive approach results in significantly faster overall execution times and higher hardware utilization than alternative approaches. A dynamic resource management is therefore mandatory for a more efficient execution of scenarios similar to our simulations, e.g. several Tsunami simulations in urgent computing, to overcome strong scalability challenges in the area of HPC. The optimizations obtained by invasive migration of cores can be generalized to similar classes of algorithms with dynamic resource requirements.

show abstract

Invasive Compute Balancing for Applications with Hybrid Parallelization

Schreiber

Riesinger

Neckel

et al. 2013

2013 25th International Symposium on Computer Architecture and High Performance Computing

View full text Add to dashboard Cite

Autotuning of Adaptive Mesh Refinement PDE Solvers on Shared Memory Architectures

Cited by 4 publications

References 9 publications

An Experience Report on (Auto-)tuning of Mesh-Based PDE Solvers on Shared Memory Systems

An Experience Report on (Auto-)tuning of Mesh-Based PDE Solvers on Shared Memory Systems

Invasive Compute Balancing for Applications with Shared and Hybrid Parallelization

Invasive Compute Balancing for Applications with Hybrid Parallelization

Contact Info

Product

Resources

About