SubsidieMeestersAccelerating Datacentre performance through Memory Chips to efficiently manage the Big Data Age
UPMEM's Processing-In-Memory technology enhances server efficiency by performing calculations within memory chips, achieving up to 20x speed and 10x energy savings for Big Data and AI applications.
Projectdetails
Introduction
Datacentres have become a backbone of the modern economy. They currently consume 2% of worldwide electricity, rising to 10% by 2030.
Energy Consumption
While compute accounts for 40% of this energy, 80% of this compute energy is related to moving data between the main memory chips and the processor (CPU). This creates a structural bottleneck within server design that causes slowdown and low CPU usage.
PIM Technology
The development of ground-breaking Processing-In-Memory (PIM) technology by UPMEM prevents this data movement by performing calculations in the memory chips, where the data resides. By eliminating the need to move data off chips, PIM bypasses structural bottlenecks.
Advantages of UPMEM's Technology
The power of UPMEM's technology comes from:
- Leveraging existing industry protocols
- Utilizing chip technology and programming languages
- Designing PIM modules that fit into standard memory slots
When compared to those using conventional memory, PIM-equipped servers have proven to be:
- Up to 20x faster
- 10x more efficient
- Cheaper for Big Data and AI applications
Financiële details & Tijdlijn
Financiële details
| Subsidiebedrag | € 2.496.229 |
| Totale projectbegroting | € 3.566.042 |
Tijdlijn
| Startdatum | 1-3-2022 |
| Einddatum | 29-2-2024 |
| Subsidiejaar | 2022 |
Partners & Locaties
Projectpartners
- SAS UPMEMpenvoerder
Land(en)
Vergelijkbare projecten binnen EIC Accelerator
| Project | Regeling | Bedrag | Jaar | Actie |
|---|---|---|---|---|
Ultra-High Speed memories for unprecedented cloud-computing performanceXenergic aims to revolutionize SRAM design for IoT and high-performance applications, achieving up to 90% energy savings and 3x speed, while seeking EIC Accelerator investment for market readiness. | EIC Accelerator | € 2.499.999 | 2022 | Details |
Full qualification, testing and commercial deployment of a unique on-chip memory technology offering the highest-density embedded memory in a standard CMOS processRAAAM aims to develop and qualify Gain-Cell Random Access Memory (GCRAM) technology for nodes ≤5nm, offering a more area-efficient and cost-effective alternative to traditional SRAM. | EIC Accelerator | € 2.499.999 | 2024 | Details |
Solving the scaling challenge of the memory industry: high-speed, low-complexity and low-cost non-volatile ferroelectric memory (Fe-NVRAM) made in EUFMC aims to revolutionize the European semiconductor market by developing innovative, energy-efficient memory solutions, reducing reliance on imports and creating local jobs. | EIC Accelerator | € 2.499.999 | 2024 | Details |
Quantum-enhanced Machine LearningEqual1 aims to finalize a scalable, sustainable quantum processor chip for AI applications, enhancing machine learning capabilities while reducing carbon footprint. | EIC Accelerator | € 2.500.000 | 2022 | Details |
Scalable Unified Processor Enhancing Revolutionary Computing, Harnessing Integrated Performance for Edge AI, Autonomous Driving, Generative AI, and Decentralized AIoT ApplicationsDeveloping the Tyr chip to enable real-time, efficient processing for Level 4/5 autonomous driving, addressing current data and processing limitations in the industry. | EIC Accelerator | € 2.499.999 | 2023 | Details |
Ultra-High Speed memories for unprecedented cloud-computing performance
Xenergic aims to revolutionize SRAM design for IoT and high-performance applications, achieving up to 90% energy savings and 3x speed, while seeking EIC Accelerator investment for market readiness.
Full qualification, testing and commercial deployment of a unique on-chip memory technology offering the highest-density embedded memory in a standard CMOS process
RAAAM aims to develop and qualify Gain-Cell Random Access Memory (GCRAM) technology for nodes ≤5nm, offering a more area-efficient and cost-effective alternative to traditional SRAM.
Solving the scaling challenge of the memory industry: high-speed, low-complexity and low-cost non-volatile ferroelectric memory (Fe-NVRAM) made in EU
FMC aims to revolutionize the European semiconductor market by developing innovative, energy-efficient memory solutions, reducing reliance on imports and creating local jobs.
Quantum-enhanced Machine Learning
Equal1 aims to finalize a scalable, sustainable quantum processor chip for AI applications, enhancing machine learning capabilities while reducing carbon footprint.
Scalable Unified Processor Enhancing Revolutionary Computing, Harnessing Integrated Performance for Edge AI, Autonomous Driving, Generative AI, and Decentralized AIoT Applications
Developing the Tyr chip to enable real-time, efficient processing for Level 4/5 autonomous driving, addressing current data and processing limitations in the industry.
Vergelijkbare projecten uit andere regelingen
| Project | Regeling | Bedrag | Jaar | Actie |
|---|---|---|---|---|
Accelerating Relational Databases with Real Memristive Processing-in-MemoryThis project aims to enhance relational database analysis speed and energy efficiency by developing a memristive memory processing unit using processing-in-memory techniques. | ERC Proof of... | € 150.000 | 2025 | Details |
Real Processing in Phase Change MemoryThe project aims to develop and commercialize a memristive memory processing unit (mMPU) using phase change memory to enhance computer performance and energy efficiency for various applications. | ERC Proof of... | € 150.000 | 2022 | Details |
ANalogue In-Memory computing with Advanced device TEchnologyThe project aims to develop closed-loop in-memory computing (CL-IMC) technology to significantly reduce energy consumption in data processing while maintaining high computational efficiency. | ERC Advanced... | € 2.498.868 | 2023 | Details |
Processing-in-memory architectures and programming libraries for bioinformatics algorithmsThis project aims to enhance genomics research by developing energy-efficient, cost-effective edge computing solutions using processing-in-memory technologies for high-throughput sequencing data analysis. | EIC Pathfinder | € 1.966.665 | 2022 | Details |
n-ary spintronics-based edge computing co-processor for artificial intelligenceMultiSpin.AI aims to revolutionize edge computing by developing a neuromorphic AI co-processor that enhances energy efficiency and processing speed, enabling transformative applications while reducing CO2 emissions. | EIC Pathfinder | € 3.143.276 | 2024 | Details |
Accelerating Relational Databases with Real Memristive Processing-in-Memory
This project aims to enhance relational database analysis speed and energy efficiency by developing a memristive memory processing unit using processing-in-memory techniques.
Real Processing in Phase Change Memory
The project aims to develop and commercialize a memristive memory processing unit (mMPU) using phase change memory to enhance computer performance and energy efficiency for various applications.
ANalogue In-Memory computing with Advanced device TEchnology
The project aims to develop closed-loop in-memory computing (CL-IMC) technology to significantly reduce energy consumption in data processing while maintaining high computational efficiency.
Processing-in-memory architectures and programming libraries for bioinformatics algorithms
This project aims to enhance genomics research by developing energy-efficient, cost-effective edge computing solutions using processing-in-memory technologies for high-throughput sequencing data analysis.
n-ary spintronics-based edge computing co-processor for artificial intelligence
MultiSpin.AI aims to revolutionize edge computing by developing a neuromorphic AI co-processor that enhances energy efficiency and processing speed, enabling transformative applications while reducing CO2 emissions.