Mobile Semiconductor's Enhanced Memory Compilers Dramatically Improve Power On Edge AI Devices
July 23, 2020 -- Mobile Semiconductor, a Seattle WA based company, today announced a new generation of Ultra Low Leakage (ULL) and Ultra Low Power (ULP) SRAM Memory Compilers that dramatically improve the performance to power ratio for Edge AI chips
Cameron Fisher, CEO and Founder of Mobile Semiconductor, said, “Drawing on the expertise we acquired developing our successful Compilers on GF’s 22FDX platform, we fully expect that these new generations of Ultra Low Leakage and Ultra Low Power Memory Compilers will meet and exceed our customer’s expectations in Edge AI.”
These new ULL and ULP Memory Compliers on GF’s 22FDX platform are unique in the industry and come with the same high-quality design and testing that customers have come to expect from Mobile Semiconductor. With leakage numbers in the nano-amp range, memory designed using Mobile Semiconductor’s compilers lead the industry.
The ULL and ULP Compilers improve the performance to power ration making them ideal for Edge AI and Machine Learning applications which are measured in tera-operations per watt (TOPS/WATT) and host new power modes giving the engineer maximum flexibility.
The graph above shows the relative power vs performance relationship between the ULP/ULL compilers optimized for Edge applications and the Ultra High Speed memory compiler which is optimized for cloud-based servers.
Fisher continued, “Mobile Semiconductor remains the leader in providing low power solutions and these compliers remove the roadblocks that engineers have struggled with in past designs. Almost every edge product developed for Edge AI, Machine Learning and IoT will have new and more stringent power demands placed on them. We believe that Mobile Semiconductor is meeting the low power needs of this market.”
About Mobile Semiconductor:
Mobile Semiconductor is headquartered Seattle, Washington with a design center in Williston, Vermont. The company develops SRAM, ROM and Register File compilers optimized for applications requiring ultra-low power, low leakage or ultra-high performance. Mobile Semiconductor’s customers differentiate their products by using our application-optimized SRAMs to meet their high performance and ultra-low power product requirements. For more information: http://www.mobile-semiconductor.com
Mobile Semiconductor is part of GLOBALFOUNDRIES’ FDXceleratorTM Partner Program and offers foundry sponsored compilers for 55nm and 28nm.
|
Related News
- X-Silicon Introduces the World's First Vulkan Driver Implementation for RISC-V, Enabling an entire Ecosystem of 3D Graphics, AI and Compute for Low-Power, Mobile, Edge and IOT Devices
- Mobile Semiconductor's 22nm ULL Memory compiler Joins the GLOBALFOUNDRIES FDXcelerator Partner program
- Samplify's APAX Technology Cuts Memory Power Consumption by 59% During Video Playback on Mobile Devices
- Dolphin Design announces partnership with Fraunhofer IIS to enhance power management in edge AI
- BrainChip & MosChip to Demonstrate Capabilities of Neural Processor IP & ASICs for Smart Edge Devices at IESA AI Summit
Breaking News
- Breker RISC-V SystemVIP Deployed across 15 Commercial RISC-V Projects for Advanced Core and SoC Verification
- Veriest Solutions Strengthens North American Presence at DVCon US 2025
- Intel in advanced talks to sell Altera to Silverlake
- Logic Fruit Technologies to Showcase Innovations at Embedded World Europe 2025
- S2C Teams Up with Arm, Xylon, and ZC Technology to Drive Software-Defined Vehicle Evolution
Most Popular
- Intel in advanced talks to sell Altera to Silverlake
- Arteris Revolutionizes Semiconductor Design with FlexGen - Smart Network-on-Chip IP Delivering Unprecedented Productivity Improvements and Quality of Results
- RaiderChip NPU for LLM at the Edge supports DeepSeek-R1 reasoning models
- YorChip announces Low latency 100G ULTRA Ethernet ready MAC/PCS IP for Edge AI
- AccelerComm® announces 5G NR NTN Physical Layer Solution that delivers over 6Gbps, 128 beams and 4,096 user connections per chipset
![]() |
E-mail This Article | ![]() |
![]() |
Printer-Friendly Page |