[go: up one dir, main page]

Zheng et al., 2022 - Google Patents

AStitch: enabling a new multi-dimensional optimization space for memory-intensive ML training and inference on modern SIMT architectures

Zheng et al., 2022

View PDF
Document ID
13439155535787267700
Author
Zheng Z
Yang X
Zhao P
Long G
Zhu K
Zhu F
Zhao W
Liu X
Yang J
Zhai J
Song S
Lin W
Publication year
Publication venue
Proceedings of the 27th ACM International Conference on Architectural Support for Programming Languages and Operating Systems

External Links

Snippet

This work reveals that memory-intensive computation is a rising performance-critical factor in recent machine learning models. Due to a unique set of new challenges, existing ML optimizing compilers cannot perform efficient fusion under complex two-level dependencies …
Continue reading at jamesthez.github.io (PDF) (other versions)

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING; COUNTING
    • G06F—ELECTRICAL DIGITAL DATA PROCESSING
    • G06F8/00—Arrangements for software engineering
    • G06F8/40—Transformations of program code
    • G06F8/41—Compilation
    • G06F8/44—Encoding
    • G06F8/443—Optimisation
    • G06F8/4441—Reducing the execution time required by the program code
    • G06F8/4442—Reducing the number of cache misses; Data prefetching
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING; COUNTING
    • G06F—ELECTRICAL DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for programme control, e.g. control unit
    • G06F9/06—Arrangements for programme control, e.g. control unit using stored programme, i.e. using internal store of processing equipment to receive and retain programme
    • G06F9/46—Multiprogramming arrangements
    • G06F9/48—Programme initiating; Programme switching, e.g. by interrupt
    • G06F9/4806—Task transfer initiation or dispatching
    • G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
    • G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING; COUNTING
    • G06F—ELECTRICAL DIGITAL DATA PROCESSING
    • G06F8/00—Arrangements for software engineering
    • G06F8/40—Transformations of program code
    • G06F8/41—Compilation
    • G06F8/45—Exploiting coarse grain parallelism in compilation, i.e. parallelism between groups of instructions
    • G06F8/456—Parallelism detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING; COUNTING
    • G06F—ELECTRICAL DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for programme control, e.g. control unit
    • G06F9/06—Arrangements for programme control, e.g. control unit using stored programme, i.e. using internal store of processing equipment to receive and retain programme
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5061—Partitioning or combining of resources
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING; COUNTING
    • G06F—ELECTRICAL DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for programme control, e.g. control unit
    • G06F9/06—Arrangements for programme control, e.g. control unit using stored programme, i.e. using internal store of processing equipment to receive and retain programme
    • G06F9/44—Arrangements for executing specific programmes
    • G06F9/455—Emulation; Software simulation, i.e. virtualisation or emulation of application or operating system execution engines
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING; COUNTING
    • G06F—ELECTRICAL DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for programme control, e.g. control unit
    • G06F9/06—Arrangements for programme control, e.g. control unit using stored programme, i.e. using internal store of processing equipment to receive and retain programme
    • G06F9/30—Arrangements for executing machine-instructions, e.g. instruction decode
    • G06F9/38—Concurrent instruction execution, e.g. pipeline, look ahead
    • G06F9/3885—Concurrent instruction execution, e.g. pipeline, look ahead using a plurality of independent parallel functional units
    • G06F9/3889—Concurrent instruction execution, e.g. pipeline, look ahead using a plurality of independent parallel functional units controlled by multiple instructions, e.g. MIMD, decoupled access or execute
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING; COUNTING
    • G06F—ELECTRICAL DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/30—Information retrieval; Database structures therefor; File system structures therefor
    • G06F17/30286—Information retrieval; Database structures therefor; File system structures therefor in structured data stores
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING; COUNTING
    • G06F—ELECTRICAL DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/36—Preventing errors by testing or debugging software
    • G06F11/3668—Software testing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING; COUNTING
    • G06F—ELECTRICAL DIGITAL DATA PROCESSING
    • G06F8/00—Arrangements for software engineering
    • G06F8/30—Creation or generation of source code
    • G06F8/31—Programming languages or programming paradigms
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING; COUNTING
    • G06F—ELECTRICAL DIGITAL DATA PROCESSING
    • G06F8/00—Arrangements for software engineering
    • G06F8/70—Software maintenance or management
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING; COUNTING
    • G06F—ELECTRICAL DIGITAL DATA PROCESSING
    • G06F8/00—Arrangements for software engineering
    • G06F8/60—Software deployment
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING; COUNTING
    • G06F—ELECTRICAL DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/50—Computer-aided design
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING; COUNTING
    • G06N—COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N99/00—Subject matter not provided for in other groups of this subclass
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING; COUNTING
    • G06N—COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computer systems utilising knowledge based models
    • G06N5/02—Knowledge representation
    • G06N5/022—Knowledge engineering, knowledge acquisition

Similar Documents

Publication Publication Date Title
Zheng et al. AStitch: enabling a new multi-dimensional optimization space for memory-intensive ML training and inference on modern SIMT architectures
Gao et al. Estimating GPU memory consumption of deep learning models
Xin et al. Accelerating human-in-the-loop machine learning: Challenges and opportunities
Jia et al. TASO: optimizing deep learning computation with automatic generation of graph substitutions
Dai et al. Reveal training performance mystery between TensorFlow and PyTorch in the single GPU environment
Pouchet et al. Loop transformations: convexity, pruning and optimization
Rasch et al. ATF: A generic directive‐based auto‐tuning framework
Zheng et al. Fusionstitching: boosting memory intensive computations for deep learning workloads
Cao et al. Moe-lightning: High-throughput moe inference on memory-constrained gpus
Rausch et al. A data-centric optimization framework for machine learning
Liou et al. GEVO: GPU code optimization using evolutionary computation
Totoni et al. HPAT: high performance analytics with scripting ease-of-use
Huang et al. Mind the gap: Attainable data movement and operational intensity bounds for tensor algorithms
Chen et al. Slapo: A schedule language for progressive optimization of large deep learning model training
Hu et al. Optimal kernel orchestration for tensor programs with korch
Fang et al. Aristotle: A performance impact indicator for the OpenCL kernels using local memory
Yu et al. Lorien: Efficient deep learning workloads delivery
Mastoras et al. Design and implementation for nonblocking execution in GraphBLAS: Tradeoffs and performance
CN118747072B (en) Compilation method, electronic device and storage medium
Ivanov et al. Sten: Productive and efficient sparsity in pytorch
Kim et al. {STRADS-AP}: Simplifying Distributed Machine Learning Programming without Introducing a New Programming Model
Kang et al. PreScaler: An efficient system-aware precision scaling framework on heterogeneous systems
Dorozhinskii et al. Fused GEMMs towards an efficient GPU implementation of the ADER‐DG method in SeisSol
US11573777B2 (en) Method and apparatus for enabling autonomous acceleration of dataflow AI applications
Alemany et al. Jespipe: a plugin-based, open MPI framework for adversarial machine learning analysis