Memory for Automotive 2022

Glossary

Definitions

Report objectives

Scope of the report

Methodology

About the authors

Companies cited in this report

3-page summary

Executive summary

Context

• Overview of the stand-alone memory market
• Overview of the automotive semicon. business

Automotive memory market forecast

• Automotive memory (all technologies)
• Focus on automotive DRAM
• Focus on automotive NAND Flash
• Focus on automotive NOR Flash

Automotive application domains

• ADAS and AD
• Digital cockpit
• Other application domains:
  • Powertrain
  • Chassis and safety
  • Body and comfort

Automotive memory technology trends

• Working Memory (DRAM)
• Data Storage (NAND)
• Code Storage (e.g., NOR)
• Other memory technologies
  • EPROM, (NV)SRAM and others

Automotive architectures and technology trend

• Software-defined vehicle
• Centralized vehicle architecture

Automotive memory players

Semiconductor supply chain – Focus on automotive Memory

Mergers & acquisitions and industry news.

General conclusions

Appendix – Teardown examples by Yole SystemPlus

About Yole Group

With a 20% CAGR_21-27, the automotive memory market is among the fastest-growing semiconductor segments.

In 2021, stand-alone memory ($167B) represented 28% of the total semiconductor market. Comparatively, the automotive memory market ($4.3B in 2021), at 2.6% of global memory market revenue, represented 10% of automotive semiconductors – showing the prevalence of non-memory electronic components in current vehicles.

We forecast memory to triple in revenue and reach 17% of the automotive semiconductor market by 2027, a 20% CAGR21-27, outperforming the global memory market (8% CAGR21-27) and the automotive semiconductor market (10% CAGR21-27).

The overall automotive memory market is currently dominated by Micron, with an estimated market share of 45% (by revenue), followed by Samsung with 13%. Infineon, Kioxia, SK Hynix and ISSI are all behind Samsung, with a market share ≤7%.

Cockpit and ADAS & autonomous driving (AD) represent more than 90% of automotive memory revenue.

In the global memory market, NAND and DRAM accounted together for ~96% of the 2021 revenue, while NOR flash accounted for only slightly more than 2% (~$3.5B). NAND and DRAM also dominated the automotive memory market, with a combined share of 80% (41% for DRAM and 39% for NAND). NOR flash has a much stronger presence in automotive, with a market share of 15% (~$0.66B).

Cockpit, with the main infotainment unit, instrument cluster, and connectivity, is currently the main memory user.  Aiming to reproduce a “smartphone-like” user experience, the cockpit memory content is following the same memory content trend. NOR flash is still being used, but managed NAND and DRAM represent most of the revenue.

ADAS & AD arrives as the second memory user in-vehicle, with 24% of the revenue in 2021. The memory consumption was mainly DRAM and high-density NOR flash, and possibly SLC NAND for smart sensors. It is the fastest-growing application domain for memory.

“Other” domain (powertrain, chassis & safety, and body & comfort) is estimated to represent about 5% of the demand. These domains, having the most constraints required, mainly use robust memory technologies such as EEPROM and NOR flash.

By 2027, Cockpit is still expected to remain the main application domain memory consumer, but ADAS & AD will see its share of the revenue increase to 36%. Technology wise, DRAM and NAND are forecasted to weigh almost 90% of automotive memory revenue.

Technology trend: smart functionalities call for faster interfaces and larger densities.

The automotive market has specific requirements in terms of quality, qualification, reliability, functional safety, and supply longevity, requiring extra effort and dedication from the players addressing this market.

For a long time, the memory technologies used were limited to robust solutions such as EEPROM and NOR flash. This has evolved with the penetration of digital cockpit and ADAS smart sensors and autonomous driving functionalities.

From analog dashboard with a main infotainment unit, vehicles are now adopting fully centralised digital cockpit electronics. DRAM, for instance, has been evolving from DDR2 and DDR3L to LPDDR4(x) and even GDDRx in some cases. An increase in associated data-storage requirements is leading to the adoption of larger and faster-managed NAND solutions. From eMMC, more and more designs are adopting UFS and potentially PCIe Solid State Drives (SSDs) in the most luxurious vehicles.

ADAS functionalities are leading to the penetration of smart sensors such as frontal camera, imaging radar, and even LiDAR. This would require high-density NOR flash – (Q)SPI to xSPI – and DRAM – DDR3L or LPDDR4 – depending on the Application Processing Unit used.

Autonomous driving requires the adoption of central processing and AI functionalities. The latter require high-bandwidth DRAM. Though most of the designs use LPDDR4(x) or DDR4, some of them are using GDDRx, and we may also think of HBM in the future. Autonomous driving requires the storage of large codes and data, necessitating the adoption of eMMC or UFS devices. Future event-data recorders (black boxes) for autonomous vehicles are forecasted to require very large densities and fast speed, which may lead to the adoption of PCIe SSDs.

"The "Memory for Automotive 2022" report by Yole provides an in-depth analysis of key technology & market trends and highlights all the challenges and opportunities foreseen in the automotive memory business.

Amazon, Ambarella,AMD, Analog Devices, Anhui Jianghuai Automotive Group, Apple, Applied Materials, Aptiv, ARM, Atlatec,Audi, AUO, Avalanche, Baidu, BMW Group, Bosch, BYD Auto,CARIAD, Carmera, Changan Automobile Group, Chery Automobile, China FAW Group Corp., Continental, CXMT, Cypress-Infineon, Denso, Dialog Semiconductor, Dongbu HiTek, Dongfeng Motor Corp.,Etron, Everspin, FCA, Ford Group, Fujitsu, GigaDevice, GlobalFoundries, GM Group, Google, Great Wall Motor Company Ltd., Guangzhou Automobile Group, Hamamatsu, H-Grace, Hitachi, HLMC, Honda, Honeywell, Huahong, Huawei, Hyundai Kia Automotive Group, Ibeo, Infineon, Intel, ISSI, Isuzu, ItoM, Kioxia, KLA, Lam Research, Level 5, Lyft , Macronix, Magna, MaxLinear, Mazda, Mediatek, Mercedes Group, Meta,MG, Microchip, Micron, Mobileye, Nanya, Nuvoton, Nvidia, NXP, OmniVision, ON Semiconductor, Optimus Ride, Panasonic, Phison , Powerchip, PSA, PSMC, Qualcomm, Rambus, Renault-Nissan-Mitsubishi Alliance, Renesas, Rohm, SAIC Motor, Samsung, SanDisk, Schneider Electric, Seagate, Siemens, Silicon motion, SK Hynix,Skyhigh, SMIC, Solidigm, Sony,Stellantis, STMicroelectronics, Subaru, Suzuki, Tata Group, TDK,Telechips, Tesla, Texas Instruments, TI, Toshiba, TowerJazz, TowerSemi, Toyota Group, TSMC, Uber, UMC, Valeo, Veoneer, VW Group, Waymo, Western Digital, Winbond, Woven Planet, XFab, Xilinx, XMC, Xperi, YMTC, Zhejiang Geely Holding Group and more

Key Features

• Memory usage, by application domain
• Market forecast 2021 - 2027 for DRAM, NAND, and NOR, by application domain
  • ADAS and AD
  • Digital cockpit
  • Other: powertrain, chassis & safety, body & comfort
• Memory use-cases, by technology
  • Code storage
  • Working memory
  • Data storage
  • Others
• Market forecast 2021 - 2027 for interface, by memory technology
  • DRAM (DDRx; LPDDRx; GDDRx; HBM)
  • NAND flash (NVMe PCIe; UFS; eMMC; SD; parallel & serial SLC NAND)
  • NOR flash ((Q)SPI; xSPI; parallel)
• Automotive memory technologies and associated players
• Player and geographical market share

Product objectives

• Provide an overview of memory technologies and markets, with a focus on automotive:
  • DRAM, NAND, NOR flash, and other (EEPROM, NVSRAM, FRAM) stand-alone memory technologies
  • Use-cases
  • Revenues, bit shipments, and volumes of device-ICs and wafers for DRAM, NAND, and NOR flash technologies – with a breakdown by interface and by application
• Present technology trends and 2021 - 2027 forecast in revenue, bits, and units (# of IC devices), by application domains:
  • Cockpit (infotainment and connectivity)
  • ADAS and autonomous driving (AD) (including event-recording systems)
  • Other: powertrain, chassis, body…
• Describe the memory use-cases and associated memories:
  • Working memory
  • Code storage
  • Data storage