Welcome
We are a design services firm specializing in cutting-edge semiconductor solutions. With a relentless commitment to innovation and a passion for pushing the boundaries of technology, we have established ourselves as a trusted partner for companies seeking top-tier semiconductor design and development expertise.
Services
GMS specializes in Custom Analog, Digital & Mixed Signal design.
With deep roots in Memory Design (DRAM, SRAM, Emerging NVM) and by extension in Compute-In-Memory architectures,
GMS is a one-stop-shop for implementing complex functionalities and algorithms into proven silicon.
Markets
GMS typically works in highly demanding and challenging markets targeting optimized Small Weight and Power (SWaP) performances.
Whitepapers
GMS is active in the community through the publication of core white papers, on its own or in partnership with its systems-level clients.
About
Our story started as the project of three friends and colleagues. It evolved into a highly professional design organization, staffed by a large crew of sound experts, and constituting a strong workforce to tackle challenging use cases.
Custom ASIC Design Services
GMS specializes in Custom Analog, Digital & Mixed Signal design. GMS engineers have deep ties to Memory Design (DRAM, SRAM, Emerging NVM) and collectively taped-out a high number of DRAM circuits.
This unique capability is combined with a know-how in logic design, both for high speed interfaces but also for general purpose full custom ASIC development.
AI Circuit Design
GMS has a long history in memory design and has been working with In-memory computing since 2015.
In-Memory Computing
Green Mountain Semiconductor has a long history in memory design, both from within the company itself and from prior work experience of our team.
Previous Experience
GMS has been doing fundamental research on compute-in-memory architectures since 2016 and filed 7 patents so far on this topic. GMS is focusing on ultra-low power in-memory neural networks for autonomous AI inferencing. This led to several development contracts with NSF and more recently with NASA. GMS is currently sponsored by this agency for designing a family of rad-hard neuromorphic processors.
Patented Technology
Green Mountain Semiconductor has several patents granted and pending in the field of commodity memory architecture, error correction code and in-memory high-parallel data processing.
Analog & Mixed Signal Design
While we also use state of the art digital tool flows where appropriate, the real world is still analog.
Analog Design
While we also use state of the art digital tool flows where appropriate, the real world is still analog. Some of our products need carefully regulated internal voltages and charge pumps along with sophisticated level shifters, others need programmable H-bridge current drivers and low-noise differential amplifiers.
Full Custom Development
Our R&D team is experienced in designing ASICs from the early technology development up to the test chips and associated test programs for functionality / yield assessment & monitoring.
Special Purpose Applications
We develop advanced concepts, IP & Asics for special applications that cannot reuse commercial off the shelf solutions.
Special Purpose Applications
Green Mountain Semiconductor is 100% US based. All of our designs are performed by our own employees at our offices in Burlington, VT.
Interface Solutions
GMS has the experience to develop high specific speed PHY interfaces based on its expertise around memories. It developed a family of secondary LPDDR4/4X/5 PHY as well as SPI/QPI.
High speed PHY’s
GMS has the experience to design high speed interfaces for general purpose data transfer, while adhering to the well known and well defined standards, for instance the ones specified by JEDEC. GMS has a track record in advanced technology nodes such as 7nm TSMC.
Markets
Green Mountain Semiconductor Inc. excels in offering advanced semiconductor solutions to a diverse array of industries. Our extensive expertise allows us to contribute meaningfully to sectors where precision and innovation are paramount. Whether addressing the complex needs of cutting-edge aerospace systems, advancing medical technology, or powering high-speed computing platforms, our commitment to excellence and innovation extends across various key markets.
Aerospace and Defense
Green Mountain is able to help reduce data transmission volume and vulnerabilities by developing chips processing critical sensor data in situ and at low power. This encompasses radiation hardness through homegrown design techniques adapted to radiation tolerant technology platforms.
Medical Devices
Green Mountain Semiconductor is a one-stop shop to develop low power Medical Device integrated circuits for patient monitoring and diagnosis from initial design to the final production device. We also developed a specific expertise around mimicking the behavior of the human brain through the development of neuromorphic edge computing IP.
High-Performance Computing
Green Mountain Semiconductor brings a sophisticated expertise in memory and high speed I/Os design. This allows to maximize in-memory computation, with the aim to reduce power and increase performance and leverage the high parallelism inherent in the memory architectures.
White Papers
Our firm has authored a multitude of comprehensive white papers. We invite you to explore our library. For full access and the ability to download any of our papers, please submit a request for access.
Co-design of a novel CMOS highly parallel, low-power, multi-chip neural network accelerator
Why do security cameras, sensors, and Siri use cloud servers instead of on-board computation? The lack of very low power, high-performance chips greatly limits the ability to field untethered edge devices. We present the NV-1, a new low-power ASIC AI processor that greatly accelerates parallel processing with dramatic reduction in energy consumption, via many parallel combined processor-memory units, i.e., a drastically non-von-Neumann architecture.
Opportunities and Limitations of in-Memory Multiply-and-Accumulate Arrays
In-memory computing is a promising solution to solve the memory bottleneck problem which becomes increasingly unfavorable in modern machine learning systems. In this paper, we introduce an architecture of random access memory (RAM) incorporating deep learning inference abilities.
Design and Testing Considerations of an In-Memory AI Chip
In-memory computing is a propitious solution for overcoming the memory bottleneck for future computer systems. In this work, we present the testing and validation considerations for a programmable artificial neural network (ANN) integrated within a phase change memory (PCM) chip.