SG10201610917TA - Accessing Data In Multi-Dimensional Tensors - Google Patents

Accessing Data In Multi-Dimensional Tensors

Info

Publication number
SG10201610917TA
SG10201610917TA SG10201610917TA SG10201610917TA SG10201610917TA SG 10201610917T A SG10201610917T A SG 10201610917TA SG 10201610917T A SG10201610917T A SG 10201610917TA SG 10201610917T A SG10201610917T A SG 10201610917TA SG 10201610917T A SG10201610917T A SG 10201610917TA
Authority
SG
Singapore
Prior art keywords
accessing data
dimensional tensors
tensors
dimensional
accessing
Prior art date
Application number
SG10201610917TA
Other languages
English (en)
Inventor
Dong Hyuk Woo
Andrew Everett Phelps
Original Assignee
Google Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Google Inc filed Critical Google Inc
Publication of SG10201610917TA publication Critical patent/SG10201610917TA/en

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30098—Register arrangements
    • G06F9/3012—Organisation of register space, e.g. banked or distributed register file
    • G06F9/3013—Organisation of register space, e.g. banked or distributed register file according to data content, e.g. floating-point registers, address registers
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30003—Arrangements for executing specific machine instructions
    • G06F9/30007—Arrangements for executing specific machine instructions to perform operations on data operands
    • G06F9/3001—Arithmetic instructions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30003—Arrangements for executing specific machine instructions
    • G06F9/30007—Arrangements for executing specific machine instructions to perform operations on data operands
    • G06F9/30036—Instructions to perform operations on packed data, e.g. vector, tile or matrix operations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30003—Arrangements for executing specific machine instructions
    • G06F9/3005—Arrangements for executing specific machine instructions to perform operations for flow control
    • G06F9/30054—Unconditional branch instructions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30003—Arrangements for executing specific machine instructions
    • G06F9/3005—Arrangements for executing specific machine instructions to perform operations for flow control
    • G06F9/30061—Multi-way branch instructions, e.g. CASE
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30003—Arrangements for executing specific machine instructions
    • G06F9/3005—Arrangements for executing specific machine instructions to perform operations for flow control
    • G06F9/30065—Loop control instructions; iterative instructions, e.g. LOOP, REPEAT
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30098—Register arrangements
    • G06F9/30101—Special purpose registers
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/32—Address formation of the next instruction, e.g. by incrementing the instruction counter
    • G06F9/321—Program or instruction counter, e.g. incrementing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/32—Address formation of the next instruction, e.g. by incrementing the instruction counter
    • G06F9/322—Address formation of the next instruction, e.g. by incrementing the instruction counter for non-sequential address
    • G06F9/325—Address formation of the next instruction, e.g. by incrementing the instruction counter for non-sequential address for loops, e.g. loop detection or loop counter
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/34—Addressing or accessing the instruction operand or the result ; Formation of operand address; Addressing modes
    • G06F9/345—Addressing or accessing the instruction operand or the result ; Formation of operand address; Addressing modes of multiple operands or results
    • G06F9/3455—Addressing or accessing the instruction operand or the result ; Formation of operand address; Addressing modes of multiple operands or results using stride
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/34—Addressing or accessing the instruction operand or the result ; Formation of operand address; Addressing modes
    • G06F9/355—Indexed addressing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/34—Addressing or accessing the instruction operand or the result ; Formation of operand address; Addressing modes
    • G06F9/355—Indexed addressing
    • G06F9/3555—Indexed addressing using scaling, e.g. multiplication of index
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/38—Concurrent instruction execution, e.g. pipeline or look ahead
    • G06F9/3836—Instruction issuing, e.g. dynamic instruction scheduling or out of order instruction execution
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2212/00—Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
    • G06F2212/45—Caching of specific data in cache memory
    • G06F2212/454—Vector or matrix data

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computational Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Algebra (AREA)
  • Computing Systems (AREA)
  • Databases & Information Systems (AREA)
  • Advance Control (AREA)
  • Executing Machine-Instructions (AREA)
  • Complex Calculations (AREA)
  • Memory System (AREA)
  • Memory System Of A Hierarchy Structure (AREA)
SG10201610917TA 2016-02-03 2016-12-28 Accessing Data In Multi-Dimensional Tensors SG10201610917TA (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US15/014,265 US9875104B2 (en) 2016-02-03 2016-02-03 Accessing data in multi-dimensional tensors

Publications (1)

Publication Number Publication Date
SG10201610917TA true SG10201610917TA (en) 2017-09-28

Family

ID=57860636

Family Applications (2)

Application Number Title Priority Date Filing Date
SG10202104454YA SG10202104454YA (en) 2016-02-03 2016-12-28 Accessing Data In Multi-Dimensional Tensors
SG10201610917TA SG10201610917TA (en) 2016-02-03 2016-12-28 Accessing Data In Multi-Dimensional Tensors

Family Applications Before (1)

Application Number Title Priority Date Filing Date
SG10202104454YA SG10202104454YA (en) 2016-02-03 2016-12-28 Accessing Data In Multi-Dimensional Tensors

Country Status (7)

Country Link
US (4) US9875104B2 (zh)
EP (1) EP3226121A3 (zh)
JP (3) JP6896415B2 (zh)
KR (4) KR101924500B1 (zh)
CN (2) CN107038018B (zh)
SG (2) SG10202104454YA (zh)
TW (4) TWI635390B (zh)

Families Citing this family (31)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9256480B2 (en) * 2012-07-25 2016-02-09 Mobileye Vision Technologies Ltd. Computer architecture with a hardware accumulator reset
US9875104B2 (en) 2016-02-03 2018-01-23 Google Llc Accessing data in multi-dimensional tensors
US10175980B2 (en) 2016-10-27 2019-01-08 Google Llc Neural network compute tile
US10360163B2 (en) 2016-10-27 2019-07-23 Google Llc Exploiting input data sparsity in neural network compute units
US9959498B1 (en) * 2016-10-27 2018-05-01 Google Llc Neural network instruction set architecture
US10896367B2 (en) * 2017-03-07 2021-01-19 Google Llc Depth concatenation using a matrix computation unit
US11567765B2 (en) 2017-03-20 2023-01-31 Intel Corporation Systems, methods, and apparatuses for tile load
US10684955B2 (en) 2017-04-21 2020-06-16 Micron Technology, Inc. Memory devices and methods which may facilitate tensor memory access with memory maps based on memory operations
US10248908B2 (en) * 2017-06-19 2019-04-02 Google Llc Alternative loop limits for accessing data in multi-dimensional tensors
US11275588B2 (en) 2017-07-01 2022-03-15 Intel Corporation Context save with variable save state size
US10175912B1 (en) 2017-07-05 2019-01-08 Google Llc Hardware double buffering using a special purpose computational unit
US10108538B1 (en) 2017-07-31 2018-10-23 Google Llc Accessing prologue and epilogue data
GB2567038B (en) * 2017-07-31 2019-09-25 Google Llc Accessing prologue and epilogue data
GB2568776B (en) 2017-08-11 2020-10-28 Google Llc Neural network accelerator with parameters resident on chip
US11620490B2 (en) * 2017-10-17 2023-04-04 Xilinx, Inc. Multi-layer neural network processing by a neural network accelerator using host communicated merged weights and a package of per-layer instructions
KR102027034B1 (ko) * 2017-11-24 2019-09-30 서울대학교산학협력단 병렬 처리를 통한 다차원 데이터 분석을 지원하는 장치 및 그 방법
WO2019113021A1 (en) * 2017-12-05 2019-06-13 Wave Computing, Inc. Tensor manipulation within a reconfigurable fabric using pointers
US10956315B2 (en) 2018-07-24 2021-03-23 Micron Technology, Inc. Memory devices and methods which may facilitate tensor memory access
CN109255438B (zh) * 2018-09-17 2020-07-17 地平线(上海)人工智能技术有限公司 调整张量数据的方法和装置
CN117785441A (zh) * 2018-12-06 2024-03-29 华为技术有限公司 处理数据的方法和数据处理装置
CN111324294B (zh) * 2018-12-17 2023-11-07 地平线(上海)人工智能技术有限公司 存取张量数据的方法和装置
CN109857744B (zh) * 2019-02-13 2020-03-20 上海燧原智能科技有限公司 稀疏张量计算方法、装置、设备及存储介质
CN111723917B (zh) * 2019-03-21 2022-11-01 中科寒武纪科技股份有限公司 运算方法、装置及相关产品
CN110209503B (zh) * 2019-08-01 2019-10-25 上海燧原智能科技有限公司 多维张量的规约计算方法、装置、设备及介质
US11687336B2 (en) * 2020-05-08 2023-06-27 Black Sesame Technologies Inc. Extensible multi-precision data pipeline for computing non-linear and arithmetic functions in artificial neural networks
CN111767508B (zh) * 2020-07-09 2024-02-23 地平线(上海)人工智能技术有限公司 计算机实现张量数据计算的方法、装置、介质和设备
CN111737007B (zh) * 2020-07-24 2021-06-18 北京一流科技有限公司 数据对象的分频处理系统及其方法
US11954580B2 (en) * 2020-09-16 2024-04-09 Meta Platforms, Inc. Spatial tiling of compute arrays with shared control
CN113836049B (zh) * 2021-09-17 2023-08-08 海飞科(南京)信息技术有限公司 存储器访问方法和电子装置
CN114489798B (zh) * 2022-01-25 2024-04-05 海飞科(南京)信息技术有限公司 用于确定张量元素的越界状态的方法和电子装置
US11868329B2 (en) 2022-05-20 2024-01-09 International Business Machines Corporation Multidimensional cube multivariate regression

Family Cites Families (32)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS58114245A (ja) * 1981-12-28 1983-07-07 Fujitsu Ltd インデツクス・インクリメントのベクトル化方式
JP3102027B2 (ja) 1990-11-20 2000-10-23 日本電気株式会社 ループ制御のネスティング管理機構
US5958048A (en) * 1996-08-07 1999-09-28 Elbrus International Ltd. Architectural support for software pipelining of nested loops
GB2359641B (en) 2000-02-25 2002-02-13 Siroyan Ltd Mapping circuitry and method
EP1261914B1 (en) * 2000-03-08 2005-12-07 Sun Microsystems, Inc. Processing architecture having an array bounds check capability
WO2001090888A1 (en) * 2000-05-23 2001-11-29 Theis Jean Paul A data processing system having an address generation unit with hardwired multidimensional memory indexing support
US6647484B1 (en) * 2000-09-19 2003-11-11 3 Dsp Corporation Transpose address mode in general purpose DSP processor
JP2003244190A (ja) * 2002-02-19 2003-08-29 Matsushita Electric Ind Co Ltd データフロー制御スイッチ用プロセッサ及びデータフロー制御スイッチ
US6952821B2 (en) * 2002-08-19 2005-10-04 Hewlett-Packard Development Company, L.P. Method and system for memory management optimization
US7107199B2 (en) * 2002-10-31 2006-09-12 Hewlett-Packard Development Company, L.P. Method and system for the design of pipelines of processors
US7047480B2 (en) * 2002-11-12 2006-05-16 Microsoft Corporation Matrix multiplication in a Galois field for forward error correction
US7225439B2 (en) * 2003-03-21 2007-05-29 Sun Microsystems, Inc. Combining write-barriers within an inner loop with fixed step
CN1259617C (zh) * 2003-09-09 2006-06-14 大唐微电子技术有限公司 一种加快rsa加/解密过程的方法及其模乘、模幂运算电路
US8539211B2 (en) 2005-01-25 2013-09-17 Nytell Software LLC Allocating registers for loop variables in a multi-threaded processor
US8806183B1 (en) 2006-02-01 2014-08-12 Ixys Ch Gmbh Blank bit and processor instructions employing the blank bit
US8904155B2 (en) 2006-03-17 2014-12-02 Qualcomm Incorporated Representing loop branches in a branch history register with multiple bits
US8438365B2 (en) * 2006-10-06 2013-05-07 Calos Fund Limited Liability Company Efficient data loading in a data-parallel processor
DE602007005790D1 (de) * 2007-06-26 2010-05-20 Ericsson Telefon Ab L M Datenverarbeitungseinheit für Anweisungen in geschachtelten Schleifen
US20100122066A1 (en) 2008-11-12 2010-05-13 Freescale Semiconductor, Inc. Instruction method for facilitating efficient coding and instruction fetch of loop construct
US20100274972A1 (en) 2008-11-24 2010-10-28 Boris Babayan Systems, methods, and apparatuses for parallel computing
CA2684226A1 (en) 2009-10-30 2011-04-30 Ibm Canada Limited - Ibm Canada Limitee Eleminating redundant operations for common properties using shared real registers
US20130185540A1 (en) * 2011-07-14 2013-07-18 Texas Instruments Incorporated Processor with multi-level looping vector coprocessor
CN104081341B (zh) * 2011-12-23 2017-10-27 英特尔公司 用于多维数组中的元素偏移量计算的指令
US20140181171A1 (en) 2012-12-24 2014-06-26 Pavel Dourbal Method and system for fast tensor-vector multiplication
US20140188961A1 (en) * 2012-12-27 2014-07-03 Mikhail Plotnikov Vectorization Of Collapsed Multi-Nested Loops
CN103440513B (zh) * 2013-09-17 2017-01-18 西安电子科技大学 基于稀疏非负张量分解的大脑特定视觉认知状态判定方法
US8842117B1 (en) 2014-02-13 2014-09-23 Raycast Systems, Inc. Computer hardware architecture and data structures for lookahead flags to support incoherent ray traversal
CN105260322B (zh) * 2014-06-30 2020-09-01 中兴通讯股份有限公司 一种对循环缓冲区消除循环移位的方法和装置
US11170294B2 (en) 2016-01-07 2021-11-09 Intel Corporation Hardware accelerated machine learning
US9875104B2 (en) * 2016-02-03 2018-01-23 Google Llc Accessing data in multi-dimensional tensors
CA2990709C (en) 2016-05-26 2018-09-04 The Governing Council Of The University Of Toronto Accelerator for deep neural networks
US10175980B2 (en) 2016-10-27 2019-01-08 Google Llc Neural network compute tile

Also Published As

Publication number Publication date
CN107038018B (zh) 2019-07-19
TWI772842B (zh) 2022-08-01
EP3226121A3 (en) 2018-10-31
JP2021128801A (ja) 2021-09-02
US20170220345A1 (en) 2017-08-03
JP2023059898A (ja) 2023-04-27
US20180107483A1 (en) 2018-04-19
US20190205141A1 (en) 2019-07-04
KR102596365B1 (ko) 2023-10-31
US10228947B2 (en) 2019-03-12
TW202014897A (zh) 2020-04-16
US10838724B2 (en) 2020-11-17
TWI635390B (zh) 2018-09-11
CN110457069A (zh) 2019-11-15
SG10202104454YA (en) 2021-06-29
JP7225314B2 (ja) 2023-02-20
TW201907305A (zh) 2019-02-16
KR101924500B1 (ko) 2018-12-03
EP3226121A2 (en) 2017-10-04
US9875104B2 (en) 2018-01-23
CN110457069B (zh) 2020-08-18
TWI681291B (zh) 2020-01-01
TW201729104A (zh) 2017-08-16
BR102016030825A2 (pt) 2018-10-09
CN107038018A (zh) 2017-08-11
TWI700587B (zh) 2020-08-01
KR20230004348A (ko) 2023-01-06
KR20180127945A (ko) 2018-11-30
KR20170092445A (ko) 2017-08-11
US20170220352A1 (en) 2017-08-03
TW202040359A (zh) 2020-11-01
KR20230153972A (ko) 2023-11-07
JP6896415B2 (ja) 2021-06-30
KR102472424B1 (ko) 2022-11-29
US9875100B2 (en) 2018-01-23
JP2017138964A (ja) 2017-08-10

Similar Documents

Publication Publication Date Title
SG10201610917TA (en) Accessing Data In Multi-Dimensional Tensors
HK1258732A1 (zh) 數據確證和存儲
HK1259159A1 (zh) 分布式數據集存儲和檢索
ZA202000319B (en) Data storage system with enforced fencing
GB2532842B (en) Mechanism and method for accessing data in a shared memory
SG11201700547WA (en) Data lineage summarization
SG11201608973TA (en) Data verification using access device
EP3127120A4 (en) Apparatuses and methods for comparing data patterns in memory
GB201501874D0 (en) Quorum based data processing
EP3229126A4 (en) Data reading method and device
EP3238078A4 (en) Speculative reads in buffered memory
SG10201504721WA (en) Seismic data processing
GB201620010D0 (en) Accessing databases
GB2549635B (en) Data retention probes and related methods
GB201515951D0 (en) Efficent spatital queries in large data tables
GB201404573D0 (en) Data processing techniques
PL3409083T4 (pl) Ulepszenia w centrach danych i związane z nimi
ZA201402292B (en) Storing and accessing data
TWI560718B (en) Data storage device and encoding method thereof
GB2530126B (en) Seismic Data Processing
GB2542867B (en) Data storage
GB201304866D0 (en) Processing microseismic data
GB201420173D0 (en) Context sensitive barriers in data processing
GB2527409B (en) Metadata based data alignment in data storage systems
GB2556472B (en) Data processing