News
IndustryTSMC’s grip on advanced chipmaking will face competitionIndustryCadence joins European Chip Design Platform to support semiconductor startups and SMEsMaking chipsMarc Hijink on ASML’s Culture, Supply Chain, and GeopoliticsProcessorsIntel's post-Nova Lake CPUs take shape as Razor Lake desktop and laptop variants surfaceAI chipsHow Google’s New AI Chip Software Helps It Battle NvidiaIndiaMore IC Packaging Plants Emerge In IndiaIndustryUS Tightens China AI Chip Access as Nvidia, TSMC and Semiconductor Stocks SlideDesignAxelera AI moves into HPC with TitaniaMaking chipsChinese envoy says chip equipment export curbs threaten EU-China trade thawIndiaSEMICON India 2026 Day 2 Sees 25 MoUs, Strategic Collaborations Across Semiconductor Value ChainIndustryTSMC’s grip on advanced chipmaking will face competitionIndustryCadence joins European Chip Design Platform to support semiconductor startups and SMEsMaking chipsMarc Hijink on ASML’s Culture, Supply Chain, and GeopoliticsProcessorsIntel's post-Nova Lake CPUs take shape as Razor Lake desktop and laptop variants surfaceAI chipsHow Google’s New AI Chip Software Helps It Battle NvidiaIndiaMore IC Packaging Plants Emerge In IndiaIndustryUS Tightens China AI Chip Access as Nvidia, TSMC and Semiconductor Stocks SlideDesignAxelera AI moves into HPC with TitaniaMaking chipsChinese envoy says chip equipment export curbs threaten EU-China trade thawIndiaSEMICON India 2026 Day 2 Sees 25 MoUs, Strategic Collaborations Across Semiconductor Value Chain
EveryChip
ARM Cortex A57 A53

Big.LITTLE

Heterogeneous CPU architecture pairing efficient and powerful cores

Photo: Open Grid Scheduler / Grid Engine, CC0, via Wikimedia Commons

Arm Holdings developed big.LITTLE to pair slow, efficient cores with fast, power-hungry ones. Announced in 2011, this design allows multi-core chips to adjust to computing needs dynamically, aiming for significant power savings compared to traditional clock scaling methods.

Share on WhatsApp

  • Announced by Arm Holdings in October 2011
  • Pairs power-hungry big cores with efficient LITTLE cores
  • Aims for up to 75 percent power savings in some tasks
  • Successor DynamIQ announced in May 2017

What it is

big.LITTLE is a processor design from Arm Holdings that combines two types of cores on one chip. It pairs high-performance cores, known as big, with energy-efficient cores, called LITTLE. The goal is to handle dynamic workloads while reducing power consumption. By switching tasks between these different core types, the system aims to save up to 75 percent of power for certain activities compared to using clock scaling alone. This architecture is commonly found in multi-processor systems on a chip.

How it came about

Arm announced the big.LITTLE concept in October 2011 alongside the Cortex-A7 core. This new core was designed to be architecturally compatible with the existing Cortex-A15. In late 2012, the company introduced the Cortex-A53 and Cortex-A57 cores, which also support this pairing for ARMv8-A architectures. Further developments included the Cortex-A12 in 2013 and the Cortex-A17 in early 2014, both of which can pair with the Cortex-A7 in a big.LITTLE configuration.

Inside the chip

The architecture optimizes for both speed and efficiency by leveraging different microarchitectures. Fast out-of-order CPUs consume more power due to leakage and switching activity. Simpler cores with fewer transistors and smaller caches use less power when idle or handling light loads. There are three main implementation models. Clustered switching moves all tasks between groups of identical cores. In-kernel switchers pair one big and one LITTLE core as a virtual unit. Heterogeneous multi-processing allows all physical cores to run simultaneously, assigning threads based on priority.

How it was made

This design is typically implemented within a multi-processor system on a chip. Manufacturers integrate the specific core designs provided by Arm into their silicon layouts. The physical arrangement depends on the chosen migration model, such as clustered pairs or asymmetric groups. For example, a chip might contain two powerful cores and four efficient ones. The interconnects, such as the Cache Coherent Interconnect, manage data transfer between the different core clusters during state migrations.

Big.LITTLE
A drawing made by EveryChip from what this page says. It is a diagram, not a picture of the real chip.

Where it went

Samsung utilized this technology in its Exynos 5 Octa series, starting with the 5410 model using clustered switching. Later versions like the 5420 adopted heterogeneous multi-processing. Nvidia employed a similar concept with a companion core in the Tegra 3 chip. Apple integrated this approach in its A series processors beginning with the A11. The Rockchip RK3399 also uses a hexa-core configuration based on this architecture. These chips are common in smartphones and tablets where battery life is critical.

Versions and family

The initial implementation relied on clustered switching, where the operating system sees only one active cluster at a time. The in-kernel switcher improved this by making paired cores visible as virtual units, allowing faster transitions via the cpufreq framework. Heterogeneous multi-processing represents the most advanced variant, enabling simultaneous use of all cores. This model supports non-symmetrical core counts and reduces kernel overhead by migrating tasks directly. Linux kernel updates, including Energy Aware Scheduling in version 5.0, have enhanced support for these varied configurations.

Why it matters

big.LITTLE addressed the need for better power management in mobile devices without sacrificing peak performance. It allowed manufacturers to build chips that could handle intensive tasks briefly while maintaining long battery life for background processes. The design influenced how operating systems schedule tasks on heterogeneous hardware. In May 2017, Arm announced DynamIQ as its successor. DynamIQ offers greater flexibility, supporting more cores per cluster and varying core designs within a single cluster, though it is incompatible with earlier big.LITTLE implementations.

Problems and reception

Practical implementations revealed some inflexibility in early models. Issues arose from mismatches between power domains and operating system features. Assigning software tasks to the correct core type proved difficult when cores had unequal abilities. Developers worked to make the electronics and software more adaptable to these challenges. The transition to heterogeneous multi-processing helped resolve many scheduling inefficiencies by exposing all cores to the kernel.

Questions people ask about Big.LITTLE

What is the main purpose of big.LITTLE?

It couples slower, battery-saving cores with powerful, power-hungry ones. This allows a multi-core processor to adjust to dynamic computing needs and use less power than clock scaling alone.

Which cores were first used in this architecture?

The Cortex-A7 was announced in October 2011 to pair with the Cortex-A15. They were designed to be architecturally compatible, allowing them to work together in a big.LITTLE chip.

How does heterogeneous multi-processing differ from other models?

It enables the use of all physical cores at the same time. High priority threads go to big cores while background tasks use LITTLE cores, unlike models that switch entire clusters off.

Good reading and sources

Where this page comes from

The facts on this page come from the Wikipedia article "ARM big.LITTLE", written by its contributors and used under the CC BY-SA 4.0 licence, and from its Wikidata record (CC0). EveryChip rewrote the text in its own words, with AI assistance, and compares every page with the article to make sure no passage is copied. It may be reused under the same CC BY-SA 4.0 licence with a link back.

It may contain mistakes, and figures change over time. Nothing here is investment or buying advice. If something is wrong, tell us. Last refreshed 2026-11-14.

Photo credits on this page

The small pictures above come from Wikimedia Commons under free licences. Some are cropped to fit.

The weekly digest

Every Monday: the chip of the day, new pages on EveryChip, and the conferences and anniversaries of the week ahead.