RECORD
Multi-dimensional intra-tile parallelization for memory-starved stencil computations
Distributed, Parallel, and Cluster Computing · Submitted 16 Oct 2015
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim \emph{Girih} shows substantial performance advantages and best arithmetic intensity at almost all problem sizes, especially on low-intensity stencils with variable coefficients.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
Optimizing the performance of stencil algorithms has been the subject of intense research over the last two decades. Since many stencil schemes have low arithmetic intensity, most optimizations focus on increasing the temporal data access locality, thus reducing the data traffic through the main memory interface with the ultimate goal of decoupling from this bottleneck. There are, however, only few approaches that explicitly leverage the shared cache feature of modern multicore chips. If every thread works on its private, separate cache block, the available cache space can become too small, and sufficient temporal locality may not be achieved. We propose a flexible multi-dimensional intra-tile parallelization method for stencil algorithms on multicore CPUs with a shared outer-level cache. This method leads to a significant reduction in the required cache space without adverse effects from hardware prefetching or TLB shortage. Our \emph{Girih} framework includes an auto-tuner to select optimal parameter configurations on the target hardware. We conduct performance experiments on two contemporary Intel processors and compare with the state-of-the-art stencil frameworks PLUTO and Pochoir, using four corner-case stencil schemes and a wide range of problem sizes. \emph{Girih} shows substantial performance advantages and best arithmetic intensity at almost all problem sizes, especially on low-intensity stencils with variable coefficients. We study in detail the performance behavior at varying grid size using phenomenological performance modeling. Our analysis of energy consumption reveals that our method can save energy by reduced DRAM bandwidth usage even at marginal performance gain. It is thus well suited for future architectures that will be strongly challenged by the cost of data movement, be it in terms of performance or energy consumption.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Optimizing the performance of stencil algorithms has been the subject of intense research over the last two decades. | background 1.00 | 0.04 | 0.01 | asserted 1.00 |
| s1 | Since many stencil schemes have low arithmetic intensity, most optimizations focus on increasing the temporal data access locality, thus reducing the data traffic through the main memory interface with the ultimate goal of decoupling from this bottleneck. | background 1.00 | 0.04 | 0.03 | asserted 0.98 |
| s2 | There are, however, only few approaches that explicitly leverage the shared cache feature of modern multicore chips. | gap 0.90 | 0.05 | 0.02 | asserted 1.00 |
| s3 | If every thread works on its private, separate cache block, the available cache space can become too small, and sufficient temporal locality may not be achieved. | background 0.79 | 0.07 | 0.02 | hedged possibility 0.85 |
| s4 | We propose a flexible multi-dimensional intra-tile parallelization method for stencil algorithms on multicore CPUs with a shared outer-level cache. | contribution 0.88 | 0.85 | 0.12 | asserted 0.99 |
| s5 | This method leads to a significant reduction in the required cache space without adverse effects from hardware prefetching or TLB shortage. | result 0.96 | 0.88 | 0.24 | asserted 1.00 |
| s6 | Our \emph{Girih} framework includes an auto-tuner to select optimal parameter configurations on the target hardware. | method 0.64 | 0.88 | 0.04 | asserted 0.99 |
| s7 | We conduct performance experiments on two contemporary Intel processors and compare with the state-of-the-art stencil frameworks PLUTO and Pochoir, using four corner-case stencil schemes and a wide range of problem sizes. | method 0.87 | 0.97 | 0.05 | asserted 0.79 |
| s8 | \emph{Girih} shows substantial performance advantages and best arithmetic intensity at almost all problem sizes, especially on low-intensity stencils with variable coefficients. | result 1.00 | 0.95 | 0.50 | asserted 1.00 |
| s9 | We study in detail the performance behavior at varying grid size using phenomenological performance modeling. | method 0.64 | 0.95 | 0.04 | asserted 0.92 |
| s10 | Our analysis of energy consumption reveals that our method can save energy by reduced DRAM bandwidth usage even at marginal performance gain. | result 0.97 | 0.96 | 0.13 | asserted 0.80 |
| s11 | It is thus well suited for future architectures that will be strongly challenged by the cost of data movement, be it in terms of performance or energy consumption. | interpretation 0.99 | 0.37 | 0.53 | asserted 0.84 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
Optimizing the performance of stencil algorithms has been the subject of intense research over the last two decades. Since many stencil schemes have low arithmetic intensity, most optimizations focus on increasing the temporal data access locality, thus reducing the data traffic through the main memory interface with the ultimate goal of decoupling from this bottleneck. There are, however, only few approaches that explicitly leverage the shared cache feature of modern multicore chips. If every thread works on its private, separate cache block, the available cache space can become too small, and sufficient temporal locality may not be achieved. We propose a flexible multi-dimensional intra-tile parallelization method for stencil algorithms on multicore CPUs with a shared outer-level cache. This method leads to a significant reduction in the required cache space without adverse effects from hardware prefetching or TLB shortage. Our \emph{Girih} framework includes an auto-tuner to select optimal parameter configurations on the target hardware. We conduct performance experiments on two contemporary Intel processors and compare with the state-of-the-art stencil frameworks PLUTO and Pochoir, using four corner-case stencil schemes and a wide range of problem sizes. \emph{Girih} shows substantial performance advantages and best arithmetic intensity at almost all problem sizes, especially on low-intensity stencils with variable coefficients. We study in detail the performance behavior at varying grid size using phenomenological performance modeling. Our analysis of energy consumption reveals that our method can save energy by reduced DRAM bandwidth usage even at marginal performance gain. It is thus well suited for future architectures that will be strongly challenged by the cost of data movement, be it in terms of performance or energy consumption.
Human review
How the literature uses it
1 arXiv paper cite this in 1 sentence
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.