Manycore processors are special kinds of
multi-core processor
A multi-core processor is a microprocessor on a single integrated circuit with two or more separate processing units, called cores, each of which reads and executes program instructions. The instructions are ordinary CPU instructions (suc ...
s designed for a high degree of
parallel processing, containing numerous simpler, independent
processor cores (from a few tens of cores to thousands or more). Manycore processors are used extensively in
embedded computers and
high-performance computing.
Contrast with multicore architecture
Manycore processors are distinct from
multi-core processor
A multi-core processor is a microprocessor on a single integrated circuit with two or more separate processing units, called cores, each of which reads and executes program instructions. The instructions are ordinary CPU instructions (suc ...
s in being optimized from the outset for a higher degree of
explicit parallelism, and for higher throughput (or lower power consumption) at the expense of latency and lower
single-thread performance
In computing, computer performance is the amount of useful work accomplished by a computer system. Outside of specific contexts, computer performance is estimated in terms of accuracy, efficiency and speed of executing computer program instruction ...
.
The broader category of
multi-core processor
A multi-core processor is a microprocessor on a single integrated circuit with two or more separate processing units, called cores, each of which reads and executes program instructions. The instructions are ordinary CPU instructions (suc ...
s, by contrast, are usually designed to efficiently run ''both'' parallel ''and'' serial code, and therefore place more emphasis on high single-thread performance (e.g. devoting more silicon to
out of order execution, deeper
pipelines, more
superscalar execution units, and larger, more general caches), and
shared memory
In computer science, shared memory is memory that may be simultaneously accessed by multiple programs with an intent to provide communication among them or avoid redundant copies. Shared memory is an efficient means of passing data between progr ...
. These techniques devote runtime resources toward figuring out implicit parallelism in a single thread. They are used in systems where they have evolved continuously (with backward compatibility) from single core processors. They usually have a 'few' cores (e.g. 2,4,8), and may be complemented by a manycore
accelerator (such as a
GPU) in a
heterogeneous system.
Motivation
Cache coherency
In computer architecture, cache coherence is the uniformity of shared resource data that ends up stored in multiple local caches. When clients in a system maintain caches of a common memory resource, problems may arise with incoherent data, whi ...
is an issue limiting the scaling of multicore processors. Manycore processors may bypass this with methods such as
message passing
In computer science, message passing is a technique for invoking behavior (i.e., running a program) on a computer. The invoking program sends a message to a process (which may be an actor or object) and relies on that process and its supporti ...
,
scratchpad memory,
DMA
DMA may refer to:
Arts
* ''DMA'' (magazine), a defunct dance music magazine
* Dallas Museum of Art, an art museum in Texas, US
* Danish Music Awards, an award show held in Denmark
* BT Digital Music Awards, an annual event in the UK
* Doctor of M ...
,
partitioned global address space
In computer science, partitioned global address space (PGAS) is a parallel programming model paradigm. PGAS is typified by communication operations involving a global memory address space abstraction that is logically partitioned, where a portion ...
, or read-only/non-coherent caches. A manycore processor using a
network on a chip
A network on a chip or network-on-chip (NoC or )This article uses the convention that "NoC" is pronounced . Therefore, it uses the convention "a" for the indefinite article corresponding to NoC ("a NoC"). Other sources may pronounce it as a ...
and local memories gives software the opportunity to explicitly optimise the spatial layout of tasks (e.g. as seen in tooling developed for
TrueNorth).
Manycore processors may have more in common (conceptually) with technologies originating in
high-performance computing such as
clusters and
vector processors
In computing, a vector processor or array processor is a central processing unit (CPU) that implements an instruction set where its instructions are designed to operate efficiently and effectively on large one-dimensional arrays of data called ...
.
GPUs may be considered a form of manycore processor having multiple
shader processing units, and only being suitable for highly parallel code (high throughput, but extremely poor single thread performance).
Suitable programming models
*
Message passing interface
*
OpenCL
OpenCL (Open Computing Language) is a framework for writing programs that execute across heterogeneous platforms consisting of central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), field-prog ...
or other APIs supporting
compute kernel
In computing, a compute kernel is a routine compiled for high throughput accelerators (such as graphics processing units (GPUs), digital signal processors (DSPs) or field-programmable gate arrays (FPGAs)), separate from but used by a main pro ...
s
*
Partitioned global address space
In computer science, partitioned global address space (PGAS) is a parallel programming model paradigm. PGAS is typified by communication operations involving a global memory address space abstraction that is logically partitioned, where a portion ...
*
Actor model
The actor model in computer science is a mathematical model of concurrent computation that treats ''actor'' as the universal primitive of concurrent computation. In response to a message it receives, an actor can: make local decisions, create mor ...
*
OpenMP
OpenMP (Open Multi-Processing) is an application programming interface (API) that supports multi-platform shared-memory multiprocessing programming in C, C++, and Fortran, on many platforms, instruction-set architectures and operating sy ...
*
Dataflow
In computing, dataflow is a broad concept, which has various meanings depending on the application and context. In the context of software architecture, data flow relates to stream processing or reactive programming.
Software architecture
Da ...
Classes of manycore systems
*
GPUs, which can be described as manycore
vector processors
In computing, a vector processor or array processor is a central processing unit (CPU) that implements an instruction set where its instructions are designed to operate efficiently and effectively on large one-dimensional arrays of data called ...
*
Massively parallel processor array
*
Asynchronous array of simple processors
Asynchrony is the state of not being in synchronization.
Asynchrony or asynchronous may refer to:
Electronics and computing
* Asynchrony (computer programming), the occurrence of events independent of the main program flow, and ways to deal wit ...
Specific manycore architectures
* ZettaScale
Japanese
PEZY Computing 2048-core modules
*
Xeon Phi coprocessor, which has MIC (''Many Integrated Cores'') architecture
*
Tilera
*
Adapteva Epiphany Architecture, a manycore chip using PGAS
scratchpad memory
*
Coherent Logix
Coherence, coherency, or coherent may refer to the following:
Physics
* Coherence (physics), an ideal property of waves that enables stationary (i.e. temporally and spatially constant) interference
* Coherence (units of measurement), a deri ...
hx3100 Processor, a 100-core DSP/GPP processor based on
HyperX Architecture
*
Movidius Myriad 2, a manycore
vision processing unit
A vision processing unit (VPU) is (as of 2018) an emerging class of microprocessor; it is a specific type of AI accelerator, designed to accelerate machine vision tasks.
Overview
Vision processing units are distinct from video processing un ...
(VPU)
*
Kalray
Kalray is a fabless semiconductor company.
Founded in 2008 as a spin-off of CEA French lab, with investors such as Renault–Nissan–Mitsubishi Alliance, Safran, NXP Semiconductors, CEA and Bpifrance.
Product history
The first Kalray paten ...
, a manycore
PCI-e accelerator for data-intensive tasks
*
Teraflops Research Chip, a manycore processor using message passing
*
TrueNorth, an
AI accelerator
An AI accelerator is a class of specialized hardware accelerator or computer system designed to accelerate artificial intelligence and machine learning applications, including artificial neural networks and machine vision. Typical applications ...
with a manycore network on a chip architecture
*
Green arrays
Green is the color between cyan and yellow on the visible spectrum. It is evoked by light which has a dominant wavelength of roughly 495570 nm. In subtractive color systems, used in painting and color printing, it is created by a co ...
, a manycore processor using message passing aimed at low power applications
*
Sunway SW26010, a 260-core manycore processor used in the, then top 1 supercomputer
Sunway TaihuLight
The Sunway TaihuLight ( ''Shénwēi·tàihú zhī guāng'') is a Chinese supercomputer which, , is ranked fourth in the TOP500 list, with a LINPACK benchmark rating of 93 petaflops. The name is translated as ''divine power, the light of Taihu Lak ...
**
SW52020, an improved 520-core variant of SW26010, with 512-bit SIMD (also adding support for half-precision), used in a prototype, meant for an exascale system (and in the future 10 exascale system), and according to datacenterdynamics China is rumored to already have two separate exascale systems secretly
*
Eyeriss, a manycore processor designed for running convolutional neural nets for embedded vision applications
*
Graphcore, a manycore
AI accelerator
An AI accelerator is a class of specialized hardware accelerator or computer system designed to accelerate artificial intelligence and machine learning applications, including artificial neural networks and machine vision. Typical applications ...
Specific manycore computers with 1M+ CPU cores
A number of computers built from multicore processors have one million or more individual CPU cores. Examples include:
*
Gyoukou (
Japanese: 暁光
Hepburn
Hepburn may refer to:
Surname
People with the surname Hepburn (the most famous in recent times being actresses Katharine Hepburn and Audrey Hepburn):
* Hepburn (surname)
Linguistics
* Hepburn romanization, a system for the romanization of Japa ...
: ''gyōkō'', dawn light), a
supercomputer developed by ExaScaler and
PEZY Computing, with 20,480,000 processing elements total plus the 1250 Intel Xeon D host processors.
*
SpiNNaker
A spinnaker is a sail designed specifically for sailing off the wind on courses between a reach (wind at 90° to the course) to downwind (course in the same direction as the wind). Spinnakers are constructed of lightweight fabric, usually n ...
, a massively parallel (1M CPU cores) manycore processor (ARM-based) built as part of the
Human Brain Project.
Specific computers with 5M+ CPU cores
Quite a few
supercomputers have over a million of even over 5 million CPU cores. When there are also coprocessors, e.g. GPUs used with, then those cores are not listed in the core-count, then quite a few more computers would hit those targets.
*
Frontier
A frontier is the political and geographical area near or beyond a boundary. A frontier can also be referred to as a "front". The term came from French in the 15th century, with the meaning "borderland"—the region of a country that fronts ...
*
Fugaku, a Japanese
supercomputer using
Fujitsu A64FX
The A64FX is a 64-bit ARM architecture microprocessor designed by Fujitsu. The processor is replacing the SPARC64 V as Fujitsu's processor for supercomputer applications. It powers the Fugaku supercomputer, the fastest supercomputer in the wo ...
ARM-based cores, 7,630,848 in total.
*
Sunway TaihuLight
The Sunway TaihuLight ( ''Shénwēi·tàihú zhī guāng'') is a Chinese supercomputer which, , is ranked fourth in the TOP500 list, with a LINPACK benchmark rating of 93 petaflops. The name is translated as ''divine power, the light of Taihu Lak ...
, a massively parallel (10M CPU cores) Chinese
supercomputer, once one of the fastest supercomputers in the world, using a custom manycore architecture. As of November 2018, the world's third fastest supercomputer (as ranked by the
TOP500
The TOP500 project ranks and details the 500 most powerful non- distributed computer systems in the world. The project was started in 1993 and publishes an updated list of the supercomputers twice a year. The first of these updates always coinc ...
list), the Chinese
Sunway TaihuLight
The Sunway TaihuLight ( ''Shénwēi·tàihú zhī guāng'') is a Chinese supercomputer which, , is ranked fourth in the TOP500 list, with a LINPACK benchmark rating of 93 petaflops. The name is translated as ''divine power, the light of Taihu Lak ...
, obtains its performance from 40,960
SW26010 manycore processors, each containing 256 cores.
See also
*
Multicore
A multi-core processor is a microprocessor on a single integrated circuit with two or more separate processing units, called cores, each of which reads and executes program instructions. The instructions are ordinary CPU instructions (such ...
*
Vector processor
In computing, a vector processor or array processor is a central processing unit (CPU) that implements an instruction set where its instructions are designed to operate efficiently and effectively on large one-dimensional arrays of data called ...
*
SIMD
Single instruction, multiple data (SIMD) is a type of parallel processing in Flynn's taxonomy. SIMD can be internal (part of the hardware design) and it can be directly accessible through an instruction set architecture (ISA), but it should ...
*
High-performance computing
High-performance computing (HPC) uses supercomputers and computer clusters to solve advanced computation problems.
Overview
HPC integrates systems administration (including network and security knowledge) and parallel programming into a multi ...
*
Computer cluster
A computer cluster is a set of computers that work together so that they can be viewed as a single system. Unlike grid computers, computer clusters have each node set to perform the same task, controlled and scheduled by software.
The comp ...
*
Multiprocessor system on a chip A multiprocessor system on a chip (, ' or ) is a system on a chip (SoC) which includes multiple microprocessors. As such, it is a multi-core system on a chip.
MPSoCs are usually targeted for embedded applications. It is used by platforms that co ...
*
Vision processing unit
A vision processing unit (VPU) is (as of 2018) an emerging class of microprocessor; it is a specific type of AI accelerator, designed to accelerate machine vision tasks.
Overview
Vision processing units are distinct from video processing un ...
*
Memory access pattern
*
Cache coherency
In computer architecture, cache coherence is the uniformity of shared resource data that ends up stored in multiple local caches. When clients in a system maintain caches of a common memory resource, problems may arise with incoherent data, whi ...
*
Embarrassingly parallel
In parallel computing, an embarrassingly parallel workload or problem (also called embarrassingly parallelizable, perfectly parallel, delightfully parallel or pleasingly parallel) is one where little or no effort is needed to separate the problem ...
*
Massively parallel
Massively parallel is the term for using a large number of computer processors (or separate computers) to simultaneously perform a set of coordinated computations in parallel. GPUs are massively parallel architecture with tens of thousands of t ...
*
CUDA
CUDA (or Compute Unified Device Architecture) is a parallel computing platform and application programming interface (API) that allows software to use certain types of graphics processing units (GPUs) for general purpose processing, an approach ...
References
External links
Architecting solutions for the Manycore future published on Feb 19, 2010 (more than one dead link in the slide)
Eyeriss architecture
{{Parallel computing
Computer architecture
Manycore processors
Parallel computing