fractal parallel computing, a geometry-inspired productivity booster
Credit: Unsplash/CC0 Public Domain

When trying to make a purchase with a shopping app, we may quickly browse the recommendation list while admitting that the machine does know about us—at least, it is learning to do so. As an effective emerging technology, machine learning (ML) has become pretty much pervasive with an application spectrum ranging from miscellaneous apps to supercomputing.

Dedicated ML computers are thus being developed at various scales, but their productivity is somewhat limited: the workload and development cost are largely concentrated in their software stacks, which need to be developed or reworked on an ad hoc basis to support every scaled model.

To solve the problem, researchers from the Chinese Academy of Sciences (CAS) proposed a fractal parallel computing model and published their research in Intelligent Computing on Sept. 5.

“Addressing the productivity issue, we proposed ML computers with fractal von Neumann architecture (FvNA),” said Yongwei Zhao, researcher from the State Key Lab of Processors, Institute of Computing Technology of CAS.

“Fractalness” is a borrowed geometric concept that describes the self-similar patterns applied to any scale. If a system is “fractal,” according to the researchers, it implies that the system always uses the same program regardless of the scale.

FvNA, a multilayered, parallelized von Neumann architecture, is not only fractal but also isostratal—which literally means “same across layered structures”.

That is, just the opposite of the conventional anisostratal ML computer architecture, FvNA adopts the same instruction-set architecture (ISA) for every layer. “The lower layer is fully controlled by the higher layer, thus, only the top layer is exposed to the programmer as a monolithic processor. Therefore, ML computers built with FvNA are programmable under a scale-invariant, homogeneous, and sequential view,” the researchers explained.

Although FvNA has been testified as applicable to the ML domain and capable of alleviating the programming productivity issue while functioning efficiently as its ad hoc counterparts, some problems remain to be solved. In this paper, the following three were addressed:

How could FvNA remain quite efficient with such a strict architectural constraint?Is FvNA also applicable to payloads from other domains?If so, what are the exact prerequisites?

To answer these questions, the researchers started by modeling the fractal parallel machine (FPM), an abstract parallel computer modeled from FvNA. FPM was built on Valiant’s multi-BSP, a homogeneous multilayered parallel model, with only minor extensions.

An instance of FPM is a tree structure of nested components; each component contains a memory, a processor, and child components. Components can execute fracops—the scheme of payloads on fractal parallel computing systems, such as reading some input data from the external storage, performing computation on the processor, and then writing output data to the external storage.

“Compared with Valiant’s multi-BSP, FPM minimized the parameters for simpler abstraction,” the researchers said. “What is more important, FPM puts explicit restrictions on the programming by only exposing a single processor to the programming interface. The processor is only aware of its parent component and child components, but not the global system specification.” In other words, the program never knows where it resides in the tree structure. Therefore, FPM cannot be programmed to be scale-dependent by definition.

Meanwhile, the researchers proposed two different ML-targeting FvNA architectures—the specific Cambricon-F and the universal Cambricon-FR—and illustrated the fractal programming style of FPM by running several general-purpose sample programs. The samples covered embarrassingly parallel, divide-and-conquer, and dynamic programming algorithms, all of which were demonstrated as efficiently programmable.

“We clarified that, although originally developed from the domain of ML, fractal parallel computing is fairly generally applicable,” the researchers concluded, drawing from their preliminary results that FPM, general-purpose and cost-optimal as it is, is as powerful as many fundamental parallel computing models such as BSP and alternating Turing machine. They also believed that full implementation of FPM could be handy in various scenarios, from the entire worldwide web to the micrometer-scale in vivo devices.

Still, the researchers pointed out a noteworthy discovery from this study that FPM limits the entropy of programming by applying constraints on the control pattern of the parallel computing systems. “Currently, fractal machines, such as Cambricon-F/FR, only leverage such entropy reduction to simplify software development,” they observed. “Whether energy reduction can be achieved by introducing fractal controlling into conventional parallel machines is an interesting open question.” More information: Yongwei Zhao et al, Fractal Parallel Computing, Intelligent Computing (2022). DOI: 10.34133/2022/9797623 Provided by Intelligent Computing Citation: Fractal parallel computing, a geometry-inspired productivity booster (2022, December 5) retrieved 5 December 2022 from This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.


TikTok EU ban on the table if social network doesn’t comply with new laws

TikTok is one of the most popular social networks out there. But TikTok is also a cause of concern for western governments that worry about the company’s ties to the Chinese government. TikTok can’t run on most devices the US government issues, and there has been talk of a ...

View more: TikTok EU ban on the table if social network doesn’t comply with new laws

Don’t Buy a Foldable Until Samsung Brings This Prototype to Life

Samsung Display via The Verge The world of foldable phones is surprisingly stagnant. The Galaxy Z Fold gets a tiny little upgrade every year, and rival phone brands loosely copy Samsung’s homework. But a new Samsung Display prototype called the “Flex In & Out” could turn this narrative on ...

View more: Don’t Buy a Foldable Until Samsung Brings This Prototype to Life

Best free sports streaming apps in 2023

Cutting the cord on cable television is something tons of people have done over the past five years. But that hasn’t proven to be the smartest way to continue to watch sports. Whether it comes from premium sports website subscriptions to keep tabs on your favorite players, or even fantasy ...

View more: Best free sports streaming apps in 2023

Avengers 5 might have Ant-Man in it, Quantumania star teases

The first MCU Phase 5 movie will be Ant-Man and the Wasp: Quantumania, the third installment in the Ant-Man franchise and a film with much higher stakes than the previous episodes. The sequel will deliver the MCU’s first Kang (Jonathan Majors) villain after we met a somewhat good He Who ...

View more: Avengers 5 might have Ant-Man in it, Quantumania star teases

Sharing a Netflix Account? Get Ready to Pay For It

DANIEL CONSTANTE/ Netflix is about to get serious in its efforts to eliminate freeloaders. If you share a Netflix account with family or friends outside your household, get ready to pay for it. A new “paid sharing” system could roll out starting next month, and you’ll have to pay a ...

View more: Sharing a Netflix Account? Get Ready to Pay For It

‘7 Wonders’ Board Game Gets a New ‘Edifice’ Expansion

Asmodee and Repos Production Board game lovers have a wonderful reason to celebrate today. Board game makers Asmodee and Repos Production announced their latest collaboration: 7 Wonders Edifice, an expansion to the popular board game 7 Wonders. The game launches on February 24th for $29.99. 7 Wonders: Edifice adds ...

View more: ‘7 Wonders’ Board Game Gets a New ‘Edifice’ Expansion

T-Mobile Kicks Off 2023 With Another Data Breach

r.classen / In a press release, T-Mobile confirms that it detected a data breach in its systems on January 5th. A “bad actor” managed to steal personal information (but not financial data) from around 37 million customers. This is the eighth T-Mobile data breach since 2018. The hacker ...

View more: T-Mobile Kicks Off 2023 With Another Data Breach

Apple appeals to UK competition watchdog investigation about mobile browser dominance

Apple has filed an appeal against the UK’s competition watchdog regarding its dominance of mobile browsers in the cloud gaming market, reports Reuters. The Competition and Markets Authority started investigating this dominance by the Cupertino firm and Google. Lawyers representing Apple believe the investigation should be reviewed as CMA ...

View more: Apple appeals to UK competition watchdog investigation about mobile browser dominance

Galaxy S23 Ultra release date and specs leak finally reveals everything about the new model

WhatsApp for iOS rolling out the ability to create a chat with yourself

Amazon Prime Music Unlimited changes streaming prices, now matches Apple Music

Deadpool 3 and Secret Wars to feature Fox’s X-Men, according to Marvel insider

Report: OLED iPad Pro still on track for 2024 release, 2026 for MacBook Pro

How to negotiate over practically anything

HomePod 2 praised in exclusive hands-on before launch

M2 Pro MacBook Pro Amazon preorder deal gives you $50 off

What “choice” means for millions of women post-Roe

Singapore FinTech firm Pilon secures $5.2M seed funding led by Wavemaker Partners

Capital Square Partners and Basil Technology team up for $700M tech fund in Asia

This feel-good movie about man’s best friend is dominating Netflix


Top Car News Car News