NDLI: Simplification and runtime resolution of data dependence constraints for loop transformations

Please wait, while we are loading the content...

Demystifying automata processing: GPUs, FPGAs or Micron's AP?

Novel HPC techniques to batch execution of many variable size BLAS computations on GPUs

Compile-time optimized and statically scheduled N-D convnet primitives for multi-core and many-core (Xeon Phi) CPUs

Fast segmented sort on GPUs

GraphGrind: addressing load imbalance of graph partitioning

Carpool: a bufferless on-chip network supporting adaptive multicast and hotspot alleviation

Iteration-fusing conjugate gradient

Efficient SIMD and MIMD parallelization of hash-based aggregation by conflict mitigation

Enabling scalability-sensitive speculative parallelization for FSM computations

Packet coalescing exploiting data redundancy in GPGPU architectures

HPAT: high performance analytics with scripting ease-of-use

Globally homogeneous, locally adaptive sparse matrix-vector multiplication on the GPU

Automatic topology mapping of diverse large-scale parallel applications

Way-combining directory: an adaptive and scalable low-cost coherence directory

Supporting automatic recovery in offloaded distributed programming models through MPI-3 techniques

Revisiting phased transactional memory

SPIRIT: a framework for creating distributed recursive tree applications

Dynamic scheduling for efficient hierarchical sparse matrix operations on the GPU

Simplification and runtime resolution of data dependence constraints for loop transformations

On improving performance of sparse matrix-matrix multiplication on GPUs

Design and implementation of bandwidth-aware memory placement and migration policies for heterogeneous memory systems

HiPA: history-based piecewise approximation for functions

Hardware/software cooperative caching for hybrid DRAM/NVM memory architectures

Frequent subtree mining on the automata processor: challenges and opportunities

Optimizing recursive task parallel programs

A performance analysis framework for exploiting GPU microarchitectural capability

SSDUP: a traffic-aware ssd burst buffer for HPC systems

libPRISM: an intelligent adaptation of prefetch and SMT levels

Simplification and runtime resolution of data dependence constraints for loop transformations

Content Provider	ACM Digital Library
Author	Pouchet, Louis-Noël Rastello, Fabrice Sampaio, Diogo N.
Abstract	Loop transformations such as tiling, parallelization or vectorization are essential tools in the quest for high-performance program execution. Precise data dependence analysis is required to determine whether the compiler can apply a transformation or not. In particular, current static analyses typically fail to provide precise enough dependence information when the code contains indirect memory accesses or polynomial subscript functions to index arrays. This leads to considering superfluous may-dependences between instructions that prevent many loop transformations to be applied. In this work we present a new hybrid (static/dynamic) framework that allows to overcome several limitations of purely static dependence analyses: For a given loop transformation, we statically generate a test to be evaluated at runtime. This test allows to determine whether the transformation is valid, and if so triggers the execution of the transformed code, falling back to the original code otherwise. Such test, originally constructed as a loop-based code with $O(n^{2})$ iterations (n being the number of iterations of the original loop-nest), are reduced to a loop-free test of O(1) complexity thanks to a new quantifier elimination scheme that we introduce in this paper. The precision and low overhead of our method is demonstrated over 25 kernels.
Starting Page	1
Ending Page	11
Page Count	11
File Format	PDF
ISBN	9781450350204
DOI	10.1145/3079079.3079098
Language	English
Publisher	Association for Computing Machinery (ACM)
Publisher Date	2017-06-14
Publisher Place	New York
Access Restriction	Subscribed
Subject Keyword	Quantifier elimination May-alias Polyhedral transformation Loop optimization Polynomials Hybrid dependence analysis
Content Type	Text
Resource Type	Article

Sl.	Authority	Responsibilities	Communication Details
1	Ministry of Education (GoI), Department of Higher Education	Sanctioning Authority	https://www.education.gov.in/ict-initiatives
2	Indian Institute of Technology Kharagpur	Host Institute of the Project: The host institute of the project is responsible for providing infrastructure support and hosting the project	https://www.iitkgp.ac.in
3	National Digital Library of India Office, Indian Institute of Technology Kharagpur	The administrative and infrastructural headquarters of the project	Dr. B. Sutradhar bsutra@ndl.gov.in
4	Project PI / Joint PI	Principal Investigator and Joint Principal Investigators of the project	Dr. B. Sutradhar bsutra@ndl.gov.in Prof. Saswat Chakrabarti will be added soon
5	Website/Portal (Helpdesk)	Queries regarding NDLI and its services	support@ndl.gov.in
6	Contents and Copyright Issues	Queries related to content curation and copyright issues	content@ndl.gov.in
7	National Digital Library of India Club (NDLI Club)	Queries related to NDLI Club formation, support, user awareness program, seminar/symposium, collaboration, social media, promotion, and outreach	clubsupport@ndl.gov.in
8	Digital Preservation Centre (DPC)	Assistance with digitizing and archiving copyright-free printed books	dpc@ndl.gov.in
9	IDR Setup or Support	Queries related to establishment and support of Institutional Digital Repository (IDR) and IDR workshops	idr@ndl.gov.in

POSTER: Hybrid Data Dependence Analysis for Loop Transformations

Mapping deviation: a technique to adapt or to guard loop transformation intuitions for legality

Black-box/white-box simplification and applications to quantifier elimination

Formula Simplification for Real Quantifier Elimination Using Geometric Invariance

Data-Flow/Dependence Profiling for Structured Transformations

Mapping Deviation: A Technique to Adapt or to Guard Loop Transformation Intuitions for Legality

Automatic modular abstractions for linear constraints

Offline Synthesis of Online Dependence Testing: Parametric Loop Pipelining for HLS

Mapping Deviation: A Technique to Adapt or to Guard Loop Transformation Intuitions for Legality

Simplification and runtime resolution of data dependence constraints for loop transformations

Similar Documents

POSTER: Hybrid Data Dependence Analysis for Loop Transformations

Mapping deviation: a technique to adapt or to guard loop transformation intuitions for legality

Black-box/white-box simplification and applications to quantifier elimination

Formula Simplification for Real Quantifier Elimination Using Geometric Invariance

Data-Flow/Dependence Profiling for Structured Transformations

Mapping Deviation: A Technique to Adapt or to Guard Loop Transformation Intuitions for Legality

Automatic modular abstractions for linear constraints

Offline Synthesis of Online Dependence Testing: Parametric Loop Pipelining for HLS

Mapping Deviation: A Technique to Adapt or to Guard Loop Transformation Intuitions for Legality

Simplification and runtime resolution of data dependence constraints for loop transformations