Heterogeneous System Architecture: Difference between revisions

Revision as of 07:49, 19 August 2016

Latest revision as of 05:37, 6 August 2025

(38 intermediate revisions by 27 users not shown)

Line 1:

{{Short description|Computing system}}

'''Heterogeneous System Architecture''' ('''HSA''') is a cross-vendor set of specifications that allow for the integration of [[central processing unit]]s and [[GPU|graphics processors]] on the same bus, with shared [[Main memory|memory]] and [[Task (computing)|tasks]].<ref>{{cite web |url=http://www.tomshardware.com/news/AMD-HSA-hUMA-APU,22324.html |title=AMD Unveils its Heterogeneous Uniform Memory Access (hUMA) Technology |website=Tom's Hardware |author=Tarun Iyer |date=30 April 2013}}</ref> The HSA is being developed by the [[HSA Foundation]], which includes (among many others) [[Advanced Micro Devices|AMD]] and [[ARM Holdings|ARM]]. The platform's stated aim is to reduce [[communication latency]] between CPUs, GPUs and other [[compute device]]s, and make these various devices more compatible from a programmer's perspective,<ref name="whitepaper">{{Cite report |author=George Kyriazis |date=30 August 2012 |title=Heterogeneous System Architecture: A Technical Review |url=http://amd-dev.wpengine.netdna-cdn.com/wordpress/media/2012/10/hsa10.pdf |publisher=AMD |access-date=26 May 2014 |archive-url=https://web.archive.org/web/20140328140823/http://amd-dev.wpengine.netdna-cdn.com/wordpress/media/2012/10/hsa10.pdf |archive-date=28 March 2014 |url-status=dead }}</ref>{{rp|3}}<ref name="whatis">{{cite web |title=What is Heterogeneous System Architecture (HSA)? |url=http://developer.amd.com/resources/heterogeneous-computing/what-is-heterogeneous-system-architecture-hsa/ |publisher=AMD |accessdateaccess-date=23 May 2014 |archive-url=https://web.archive.org/web/20140621213832/http://developer.amd.com/resources/heterogeneous-computing/what-is-heterogeneous-system-architecture-hsa/ |archive-date=21 June 2014 |url-status=dead }}</ref> relieving the programmer of the task of planning the moving of data between devices' disjoint memories (as must currently be done with [[OpenCL]] or [[CUDA]]).<ref>{{cite web |author=Joel Hruska |title=Setting HSAIL: AMD explains the future of CPU/GPU cooperation |url=http://www.extremetech.com/gaming/164817-setting-hsail-amd-cpu-gpu-cooperation |website=[[ExtremeTech]] |publisher=[[Ziff Davis]] |date=2013-08-26}}</ref>

CUDA and OpenCL as well as most other fairly advanced programming languages can use HSA to increase their execution performance.<ref>{{cite web|url=http://www.slideshare.net/mobile/linaroorg/hsa-linaro-updatejuly102013|title=LCE13: Heterogeneous System Architecture (HSA) on ARM|author=Linaro|work=slideshare.net|date=21 March 2014}}</ref> [[Heterogeneous computing]] is widely used in [[MPSoC|system-on-chip]] devices such as [[Tablet computer|tablets]], [[smartphone]]s, other mobile devices, and [[video game console]]s.<ref name="gpuscience">{{cite web

| url = http://gpuscience.com/cs/heterogeneous-system-architecture-purpose-and-outlook/

| archiveurlarchive-url = httphttps://web.archive.org/web/20140201183411/http://gpuscience.com/cs/heterogeneous-system-architecture-purpose-and-outlook/

| title = Heterogeneous System Architecture: Purpose and Outlook

| date = 2012-11-09 | accessdateaccess-date = 2014-05-24

| archivedatearchive-date = 2014-02-01

| website = gpuscience.com

}}</ref> HSA allows programs to use the graphics processor for [[floating point]] calculations without separate memory or scheduling.<ref>{{cite web |title=Heterogeneous system architecture: Multicore image processing using a mix of CPU and GPU elements |website=Embedded Computing Design |url=http://embedded-computing.com/articles/heterogeneous-processing-using-mix-cpu-gpu-elements/ |accessdateaccess-date=23 May 2014}}</ref>

== Rationale ==

The rationale behind HSA is to ease the burden on programmers when offloading calculations to the GPU. Originally driven solely by AMD and called the FSA, the idea was extended to encompass processing units other thenthan GPUs, e.g.such theas widespreadother manufacturers' [[digital signal processor|DSPs]] by other hardware design companies, as well.

{{Gallery

Line 17 ⟶ 18:

| height = 190

| align = center

| lines = 3

| File:HSA – using the GPU without HSA.svg

| Steps performed when offloading calculations to the [[Graphics processing unit|GPU]] on a non-HSA system

Line 26 ⟶ 25:

}}

Modern GPUs are very well suited to perform [[Singlesingle instruction, multiple data]] (SIMD) and [[Singlesingle instruction, multiple threads]] (SIMT), while modern CPUs are still being optimized for branching. etc.

== Overview ==

{{Refimprove section|date=May 2014}}

== Overview ==

{{RefimproveMore citations needed section|date=May 2014}}

Originally introduced by [[embedded system]]s such as the [[Cell Broadband Engine]], sharing system memory directly between multiple system actors makes heterogeneous computing more mainstream. Heterogeneous computing itself refers to systems that contain multiple processing units{{snd}} [[central processing unit]]s (CPUs), [[graphics processing unit]]s (GPUs), [[digital signal processor]]s (DSPs), or any type of [[application-specific integrated circuit]]s (ASICs). The system architecture allows any accelerator, for instance a [[GPU|graphics processor]], to operate at the same processing level as the system's CPU.

Among its main features, HSA defines a unified [[virtual address space]] for compute devices: where GPUs traditionally have their own memory, separate from the main (CPU) memory, HSA requires these devices to share [[Pagepage (computer memory)|page tables]] so that devices can exchange data by sharing [[Pointerpointer (computer programming)|pointers]]. This is to be supported by custom [[memory management unit]]s.<ref name="whitepaper"/>{{rp|6–7}} To render interoperability possible and also to ease various aspects of programming, HSA is intended to be [[Instructioninstruction set|ISA]]-agnostic for both CPUs and accelerators, and to support high-level programming languages.

So far, the HSA specifications cover:

* HSA Intermediate Layer (HSAIL), a [[p-code machine|virtual instruction set]] for parallel programs

===HSA Intermediate Layer===<!--incoming redirect-->

** similar{{according to whom|date=May 2015}} to [[LLVM Intermediate Representation]] and [[Standard Portable Intermediate Representation|SPIR]] (used by [[OpenCL]] and [[Vulkan (API)|Vulkan]])

*HSAIL HSA(Heterogeneous IntermediateSystem LayerArchitecture (HSAILIntermediate Language), a [[p-code machine|virtual instruction set]] for parallel programs

** finalized to a specific instruction set by a [[Just-in-time compilation|JIT compiler]]

** similar{{according to whom|date=May 2015}} to [[LLVM Intermediate Representation]] and [[Standard Portable Intermediate Representation|SPIR]] (used by [[OpenCL]] and [[Vulkan (API)|Vulkan]])

** make late decisions on which core(s) should run a task

** finalized to a specific instruction set by a [[Just-in-time compilation|JIT compiler]]

** explicitly parallel

** make late decisions on which core(s) should run a task

** supports exceptions, virtual functions and other high-level features

** explicitly parallel

** syscall methods (I/O, [[printf]],{{clarify|reason=printf is not a syscall on any operating system that I know|date=May 2015}} etc.)

** supports exceptions, virtual functions and other high-level features

** debugging support

* HSA memory model

** compatible with [[C++11]], OpenCL, [[Java (programming language)|Java]] and [[.NET Framework|.NET]] memory models

* ===HSA memory model===

** relaxed consistency

** compatible with [[C++11]], OpenCL, [[Java (programming language)|Java]] and [[.NET Framework|.NET]] memory models

** designed to support both managed languages (e.g. Java) and unmanaged languages (e.g. [[C (programming language)|C]])

** relaxed consistency

** will make it much easier to develop 3rd-party compilers for a wide range of heterogeneous products programmed in [[Fortran]], C++, [[C++ AMP]], Java, et al.

** designed to support both managed languages (e.g. Java) and unmanaged languages (e.g. [[C (programming language)|C]])

* HSA dispatcher and run-time

** will make it much easier to develop 3rd-party compilers for a wide range of heterogeneous products programmed in [[Fortran]], C++, [[C++ AMP]], Java, et al.

** designed to enable heterogeneous task queueing: a work queue per core, distribution of work into queues, load balancing by work stealing

** any core can schedule work for any other, including itself

* ===HSA dispatcher and run-time===

** significant reduction of overhead of scheduling work for a core

** designed to enable heterogeneous task queueing: a work queue per core, distribution of work into queues, load balancing by work stealing

** any core can schedule work for any other, including itself

** significant reduction of overhead of scheduling work for a core

Mobile devices are one of the HSA's application areas, in which it yields improved power efficiency.<ref name="gpuscience" />

=== Block diagrams ===

The block diagramsillustrations below providecompare highCPU-levelGPU illustrationscoordination of howunder HSA operates and how it comparesversus tounder traditional architectures.

{{Gallery

Line 63 ⟶ 64:

| height = 190

| align = center

| lines = 3

| File:Desktop computer bus bandwidths.svg

| Standard architecture with a discrete [[graphics card|GPU]] attached to the [[PCI Express]] bus. [[Zero-copy]] between the GPU and CPU is not possible due to distinct physical memories.

|File:HSA-enabled virtual memory with distinct graphics card.svg

| HSA brings unified virtual memory, and facilitates passing pointers over PCI Express instead of copying the entire data.

| File:Integrated graphics with distinct memory allocation.svg

| In partitioned main memory, one part of the system memory is exclusively allocated to the GPU. As a result, zero-copy operation areis not possible.

| File:HSA-enabled integrated graphics.svg

| Unified main memory, madewhere possibleGPU byand aCPU combination ofare HSA-enabled GPU and CPU. AsThis a result, it is possible to performmakes zero-copy operationsoperation possible.<ref>{{cite web |url=http://www.semiaccurate.com/2014/01/15/technical-look-amds-kaveri-architecture/ |title=Kaveri microarchitecture |date=2014-01-15 |work=[[SemiAccurate]]}}</ref>

| File:MMU and IOMMU.svg

| Both theThe CPU's [[Memory management unit|MMU]] and the GPU's [[IOMMU]] havemust toboth comply with the HSA hardware specifications.

}}

{{Clear}}

==Software support{{Anchor|AMDKFD|HQ|HMM}}Software support ==

[[File:Linux AMD graphics stack.svg|thumb|right|upright=1.8|AMD GPUs contain certain additional functional units intended to be used as part of HSA. In Linux, kernel driver {{Mono|amdkfd}} provides required support.<ref>{{cite web

| url = httphttps://www.phoronix.com/scan.php?page=news_item&px=MTc0NTk

| title = AMDKFD Driver Still Evolving For Open-Source HSA On Linux

| date = July 21, July 2014 | accessdateaccess-date = January 21, January 2015

| author = Michael Larabel | publisher = [[Phoronix]]

}}</ref><ref name="kernelnewbies-3.19" />]]

Some of the HSA-specific features implemented in the hardware need to be supported by the [[operating system kernel]] and specific device drivers. For example, support for AMD [[Radeon]] and [[AMD FirePro]] graphics cards, and [[AMD Accelerated Processing Unit|APUs]] based on [[Graphics Core Next]] (GCN), was merged into version 3.19 of the [[Linux kernel mainline]], released on February 8, February 2015.<ref name="kernelnewbies-3.19">{{cite web

| url = http://kernelnewbies.org/Linux_3.19#head-ae54e026ef7588f4431f7e94178d27d5cd830bbf

| title = Linux kernel 3.19, Section 1.3. HSA driver for AMD GPU devices

| date = February 8, February 2015 | accessdateaccess-date = February 12, February 2015

| website = kernelnewbies.org

}}</ref> Programs do not interact directly with {{Mono|amdkfd}}{{Explain|date=December 2023}}, but queue their jobs utilizing the HSA runtime.<ref>{{cite web

| url = https://github.com/HSAFoundation/HSA-Runtime-Reference-Source/blob/master/README.md

| title = HSA-Runtime-Reference-Source/README.md at master

| date = November 14, November 2014 | accessdateaccess-date = February 12, February 2015

| website = github.com

}}</ref> This very first implementation, known as {{Mono|amdkfd}}, focuses on [[AMD Accelerated Processing Unit#Steamroller architecture .282014.29: Kaveri|"Kaveri"]] or "Berlin" APUs and works alongside the existing Radeon kernel graphics driver.

Additionally, {{Mono|amdkfd}} supports ''heterogeneous queuing'' (HQ), which aims to simplify the distribution of computational jobs among multiple CPUs and GPUs from the programmer's perspective. {{As of|2015|2}}, supportSupport for ''heterogeneous memory management'' (''HMM''), suited only for graphics hardware featuring version 2 of the AMD's [[IOMMU]], has not yet beenwas accepted into the Linux kernel mainline version 4.14.<ref>{{cite web|url=https://www.xda-developers.com/linux-kernel-414/|archive-url=https://web.archive.org/web/20171113231202/https://www.xda-developers.com/linux-kernel-414/|url-status=dead|archive-date=13 November 2017|title=Linux Kernel 4.14 Announced with Secure Memory Encryption and More|date=13 November 2017}}</ref>

Integrated support for HSA platforms has been announced for the "Sumatra" release of [[OpenJDK]], due in 2015.<ref>{{cite web |url=http://www.hpcwire.com/2013/08/26/hsa_foundation_aims_to_boost_javas_gpu_prowess/ |title=HSA Foundation Aims to Boost Java’sJava's GPU Prowess |author=Alex Woodie |date=26 August 2013 |website=HPCwire}}</ref>

[[AMD APP SDK]] is AMD's proprietary software development kit targeting [[parallel computing]], available for Microsoft Windows and Linux. Bolt is a C++ template library optimized for heterogeneous computing.<ref>{{cite web |url=https://github.com/HSA-Libraries/Bolt |title=Bolt on github|website=[[GitHub]]|date=11 January 2022}}</ref>

[[GPUOpen]] comprehends a couple of other software tools related to HSA. [[Code XLCodeXL]] version 2.0 includes an HSA profiler.<ref>{{cite web |url=http://gpuopen.com/codexl-2-0-is-here-and-open-source/ |title=CodeXL 2.0 includes HSA profiler |author=AMD GPUOpen |date=2016-04-19 |access-date=21 April 2016 |archive-date=27 June 2018 |archive-url=https://web.archive.org/web/20180627034628/https://gpuopen.com/codexl-2-0-is-here-and-open-source/ |url-status=dead }}</ref>

{{Clear}}

== Hardware support ==

=== AMD ===

{{As of|2015|2}}, only AMD's "Kaveri" A-series APUs (cf. [[List of AMD Accelerated Processing Unit microprocessors#"Kaveri" (2014, 28 nm)|"Kaveri" desktop processors]] and [[List of AMD Accelerated Processing Unit microprocessors#"Kaveri" 2014, 28 nm|"Kaveri" mobile processors]]) and Sony's [[PlayStation 4]] allowed the [[Graphics processing unit#Integrated graphics|integrated GPU]] to access memory via version 2 of the AMD's IOMMU. Earlier APUs (Trinity and Richland) included the version 2 IOMMU functionality, but only for use by an external GPU connected via PCI Express.{{Citation needed|date=June 2016}}

=== AMD ===

{{As of|2015|2}}, only AMD's "Kaveri" A-series APUs (cf. [[List of AMD Accelerated Processing Unit microprocessors#"Kaveri" (2014, 28 nm)|"Kaveri" desktop processors]] and [[List of AMD Accelerated Processing Unit microprocessors#"Kaveri" 2014, 28 nm|"Kaveri" mobile processors]]) and Sony's [[PlayStation 4]] allowed the integrated GPU to access memory via version 2 of the AMD's IOMMU. Earlier APUs (Trinity and Richland) included the version 2 IOMMU functionality, but only for use by an external GPU connected via PCI Express.{{Citation needed|date=June 2016}}

Post-2015 Carrizo and Bristol Ridge APUs also include the version 2 IOMMU functionality for the integrated GPU.{{Citation needed|date=June 2016}}

Line 122 ⟶ 120:

{{AMD APU features}}

=== ARM ===

ARM's [[Bifrost (microarchitecture)|Bifrost]] microarchitecture, as implemented in the Mali-G71,<ref>{{cite web |url=http://www.anandtech.com/show/10375/arm-unveils-bifrost-and-mali-g71/5 |archive-url=https://archive.today/20160910101608/http://www.anandtech.com/show/10375/arm-unveils-bifrost-and-mali-g71/5 |url-status=dead |archive-date=10 September 2016 |title=ARM Bifrost GPU Architecture |date=2016-05-30}}</ref> is fully compliant with the HSA 1.1 hardware specifications. {{As of|2016|6}}, ARM has not announced software support that would use this hardware feature.

== See also ==

{{Commons category}}

== See also ==

* [[General-purpose computing on graphics processing units]] (GPGPU)

* [[Non-Uniform Memory Access]] (NUMA)

Line 133 ⟶ 129:

* [[Shared memory]]

* [[Zero-copy]]

* A technique enabling zero-copy operation for a CPU and a parallel accelerator <ref> Computer memory architecture for hybrid serial and parallel computing systems, US patents 7,707,388, 2010 and 8,145,879, 2012. Inventor: [[Uzi Vishkin]] </ref>

== References ==

{{Reflist|30em}}

== External links ==

{{Commons category}}

* {{YouTube|id=ln8JpfaLvbM|title=HSA Heterogeneous System Architecture Overview}} by Vinod Tipparaju at [[ACM/IEEE Supercomputing Conference|SC13]] in November 2013

* [https://web.archive.org/web/20160514070602/http://www.mpsoc-forum.org/previous/2013/slides/8-Hegde.pdf HSA and the software ecosystem]

* [http://www-conf.slac.stanford.edu/xldb2012/talks/xldb2012_wed_1400_MichaelHouston.pdf 2012 – HSA by Michael Houston] {{Webarchive|url=https://web.archive.org/web/20160305141652/http://www-conf.slac.stanford.edu/xldb2012/talks/xldb2012_wed_1400_MichaelHouston.pdf |date=5 March 2016 }}

{{Use dmy dates|date=July 2019}}

[[Category:Heterogeneous System Architecture| ]]