Integrating profile-driven parallelism detection and machine-learning-based mapping

Wang, Zheng; Tournavitis, Georgios; Franke, Björn; O’Boyle, Michael

doi:10.1145/2579561

Cited by 59 publications

(37 citation statements)

References 68 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Predictive Modeling Machine learning based predictive modeling is emerging as a powerful technique for optimizing parallel programs [28,29,33,37,[39][40][41]. Its great advantage is its ability to adapt to changing platforms as it has no a prior assumptions about their behavior.…”

Section: Related Workmentioning

confidence: 99%

Adaptive optimization for OpenCL programs on embedded heterogeneous systems

2017

Self Cite

View full text Add to dashboard Cite

Heterogeneous multi-core architectures consisting of CPUs and GPUs are commonplace in today's embedded systems. These architectures offer potential for energy efficient computing if the application task is mapped to the right core. Realizing such potential is challenging due to the complex and evolving nature of hardware and applications. This paper presents an automatic approach to map OPENCL kernels onto heterogeneous multi-cores for a given optimization criterion -whether it is faster runtime, lower energy consumption or a trade-off between them. This is achieved by developing a machine learning based approach to predict which processor to use to run the OPENCL kernel and the host program, and at what frequency the processor should operate. Instead of hand-tuning a model for each optimization metric, we use machine learning to develop a unified framework that first automatically learns the optimization heuristic for each metric off-line, then uses the learned knowledge to schedule OPENCL kernels at runtime based on code and runtime information of the program. We apply our approach to a set of representative OPENCL benchmarks and evaluate it on an ARM big.LITTLE mobile platform. Our approach achieves over 93% of the performance delivered by a perfect predictor. We obtain, on average, 1.2x, 1.6x, and 1.8x improvement respectively for runtime, energy consumption and the energy delay product when compared to a comparative heterogeneous-aware OPENCL task mapping scheme.

show abstract

Section: Related Workmentioning

confidence: 99%

Adaptive optimization for OpenCL programs on embedded heterogeneous systems

2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…The supervised methods usually distinguish and label the code regions which are suitable for parallelization or not, and then use the information to train the model [78]. While unsupervised learning methods are mainly based on clustering techniques and employ the historical data or profile data [79] to find program fragments with similar patterns [80].…”

Section: Intelligent Algorithmsmentioning

confidence: 99%

A Survey of Loop Parallelization: Models, Approaches, and Recent Developments

Ye¹,

Deng²,

Zou³

2016

IJGDC

View full text Add to dashboard Cite

In cloud computing era, automatic parallelization is still significant for virtualization platform. However, after several decades of development, the overall effect is still to be improved. Summary of the mainstream technology developments will be beneficial to reveal the future direction and trend. This paper reviews the technology of loop parallelization, which is the key issue in automatic parallelization. After introducing the basic models and approaches, we focus on the recent developments, on which we obtain the trend of this field and the conclusions about future.

show abstract

“…HELIX [10] uses a dynamic loop nesting graph to select a set of loops to parallelize. [30] uses dynamic analysis of control and data dependencies as input to a trained predictor to autoparallelize loops, relying on the user to check correctness.…”

Section: Generated Halide Code For Photoshop Filtersmentioning

confidence: 99%

Helium: lifting high-performance stencil kernels from stripped x86 binaries to halide DSL code

MendisCharith¹,

BosboomJeffrey²,

WuKevin³

et al. 2015

SIGPLAN Not.

View full text Add to dashboard Cite

Highly optimized programs are prone to bit rot, where performance quickly becomes suboptimal in the face of new hardware and compiler techniques. In this paper we show how to automatically lift performance-critical stencil kernels from a stripped x86 binary and generate the corresponding code in the high-level domain-specific language Halide. Using Halide's state-of-the-art optimizations targeting current hardware, we show that new optimized versions of these kernels can replace the originals to rejuvenate the application for newer hardware.The original optimized code for kernels in stripped binaries is nearly impossible to analyze statically. Instead, we rely on dynamic traces to regenerate the kernels. We perform buffer structure reconstruction to identify input, intermediate and output buffer shapes. We abstract from a forest of concrete dependency trees which contain absolute memory addresses to symbolic trees suitable for high-level code generation. This is done by canonicalizing trees, clustering them based on structure, inferring higher-dimensional buffer accesses and finally by solving a set of linear equations based on buffer accesses to lift them up to simple, high-level expressions.Helium can handle highly optimized, complex stencil kernels with input-dependent conditionals. We lift seven kernels from Adobe Photoshop giving a 75% performance improvement, four kernels from IrfanView, leading to 4.97× performance, and one stencil from the miniGMG multigrid benchmark netting a 4.25× improvement in performance. We manually rejuvenated Photoshop by replacing eleven of Photoshop's filters with our lifted implementations, giving 1.12× speedup without affecting the user experience.

show abstract

Integrating profile-driven parallelism detection and machine-learning-based mapping

Cited by 59 publications

References 68 publications

Adaptive optimization for OpenCL programs on embedded heterogeneous systems

Adaptive optimization for OpenCL programs on embedded heterogeneous systems

A Survey of Loop Parallelization: Models, Approaches, and Recent Developments

Helium: lifting high-performance stencil kernels from stripped x86 binaries to halide DSL code

Contact Info

Product

Resources

About