BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
HappyChippy
Member
A bit random, but does anyone here have LinkedIn premium and/or can potentially work out who BRNs recent followers are?
Could potentially be telling…
I have noticed the number climbing up very consistently over the last 6months to a year.
Could potentially be telling…
I have noticed the number climbing up very consistently over the last 6months to a year.
DingoBorat
Slim
Fullmoonfever
Top 20
Neuromorphic engineering and urban farming advance concurrently
Discover how neuromorphic engineering and urban farming are converging to revolutionize sustainable living and advanced technology in urban landscapes.
www.baselinemag.com
Neuromorphic engineering and urban farming advance concurrently
- By Adam Campbell
- Last updated July 2, 2024
Neuromorphic engineering is steadily advancing, modeling itself after the intricacies of the human brain.5
BrainChip Holdings, an Australian company, specializes in this unique field with their standout product, Akida. This revolutionary System-on-Chip (SoC) mirrors the human brain’s processing mechanism.
Notably, Akida reduces power consumption, making it ideal for devices reliant on extended battery life. It also boasts adaptability, easily syncing with a range of machinery from self-driving vehicles to IoT devices. All the signs point to Akida reshaping how AI systems calculate and interact with the world, ultimately pushing cognitive computing forward.
Bioprinting, while a specialized technology in the healthcare domain, harnesses 3D printing to layer biological materials, creating products with a far-reaching potential to shift paradigms in medical treatment, drug development, and food production. With an estimated market value of $1.8 billion by 2027, the bioprinting field is only expanding. This growth boils down to a combination of an escalating need for organ transplants, advancements in 3D printing technology and the increasing prevalence of chronic diseases.
In the bioprinting market, Organovo has proven itself noteworthy, developing 3D human tissues that simulate the structure and function of organic tissues.
Neuromorphic engineering and urban farming’s concurrent progress
This ground-breaking work suggests a future rich with possibilities in bioprinting.In contrast, Urban Farming Technology focuses on the practical application of urban spaces for farming. This innovative field changes the game for food production and consumption. By offering city residents easy access to home-grown vegetables, Urban Farming Technology can tackle the challenge of decreasing farmland while promoting sustainability.
Urban farming techniques such as Vertical and Hydroponic farming can yield a hundredfold increase in produce compared to conventional farming. Not only is the food organic and nutrient-dense, but the methodology also contributes to a healthier lifestyle, reduces waste, and promotes community involvement and local economies. It’s clear that Urban Farming Technology is not just a farming fad; it’s leading us toward a healthier and greener lifestyle.
Despite the lack of recognition, these sectors carry immense potential and live on the brink of ground-breaking innovation and disruption. These promising sectors are providing a rich playground for daring startups ready to tap into their potential.
Adam Campbell
The IT mastermind with a passion for data and a flair for blogging brilliance. When he's not conquering tech conundrums as an IT manager, you'll find him surfing the waves of big data, decoding the secrets of technology, and serving up witty insights with a side of code humor on his blog.
Fullmoonfever
Top 20
Preprint paper from 1st July with some data and thoughts across the spectrum for perspective.
Authors from Uni of Dayton and the AFRL.
Power consumption can vary obviously depending on end use requirements.
Paper attached also.
Survey of deep learning accelerators for edge and emerging computing
Authors from Uni of Dayton and the AFRL.
Power consumption can vary obviously depending on end use requirements.
Paper attached also.
Survey of deep learning accelerators for edge and emerging computing
Pom down under
Regular
Damn it esq, don't say things like that around here. You will get reported and deported back to Hot Crapper.
You've suspiciously written something that an MF article would say. Are you a down ramper by any chance? Well if so then you are not doing a great job at it.
Keep the vibes positive please.
Sean is our golden boy and he deserves all the monies and then some. We hired him because he has very high connections in silicon valley and we are so very close to signing multi billion dollar NDAs with with Nanose and Nintendo. There is no further need to justify why our tech will be in every household items in the world. Just like FF, I am off to spread the good news somewhere else.
Weeeeeeeeee......!!!!!!
As always,
Not advice. DYOR
Zed, this is all time best gif ever for Sean.
Someone needs to print this on a shirt and wear it to the next AGM. Hahahaha
Waiting Waiting and waiting. Will it happen, who knows loosing faithDamn it esq, don't say things like that around here. You will get reported and deported back to Hot Crapper.
You've suspiciously written something that an MF article would say. Are you a down ramper by any chance? Well if so then you are not doing a great job at it.
Keep the vibes positive please.
Sean is our golden boy and he deserves all the monies and then some. We hired him because he has very high connections in silicon valley and we are so very close to signing multi billion dollar NDAs with with Nanose and Nintendo. There is no further need to justify why our tech will be in every household items in the world. Just like FF, I am off to spread the good news somewhere else.
Weeeeeeeeee......!!!!!!
As always,
Not advice. DYOR
buena suerte :-)
BOB Bank of Brainchip
This is a piece from the 2024 AGM transcript where SH mentions...
"closing in on a decision" and .....
"Based on direct prospect feedback I believe we are well positioned in some of the more critical engagements. In a number of cases, we have made it through the down selection process that eliminates many competitors and focuses on only 2 or 3 vendors for final selection."
Last edited:
Quiltman
Regular
I know most here understand this chart : the chart of the entrepreneur.
It's difficult when you are in the midst of the action, privy to all that is going on.
But when you are on the outside looking in, as are shareholders, somewhat manic !
What Brainchip in delivering to the market is revolutionary, and IMO following this curve.
We are at that point of " Crash & Burn - ie. give up " , or moving towards the uptrend - and everyone is feeling nervous.
Like many here, I attended the AGM to try to be as informed as possible from the business, not hearsay & scuttlebutt, and I satisfied myself, along with all the data that has been made available, that we will succeed.
But I truly do understand those that are losing hope ... it is to be expected ... we are following the normal & expected path of any disruptive technology or entrepreneurial endeavour.
Pom down under
Regular
Earlyrelease
Regular
Good things come to those that wait.Me waiting for a price sensitive announcement.
Always hated that saying when mum used to say it but it’s a tried and tested historical saying for a reason.
Getupthere
Regular
Edge AI accelerator delivers 60 TOPS at minimal power — Electronic Products
EdgeCortix’s SAKURA-II energy-efficient edge AI accelerator handles the latest GenAI tasks at the edge, including LLMs and LVMs.
Edge AI accelerator delivers 60 TOPS at minimal power
Jul 3, 2024 | 7:31 AM
Startup EdgeCortix Inc. recently launched its next-generation SAKURA-II edge AI accelerator, designed for processing generative AI (GenAI) workloads at the edge with minimal power consumption and low latency. This platform is paired with the company’s second-generation dynamic neural accelerator (DNA) architecture to address the most challenging GenAI tasks in the industry.
EdgeCortix, headquartered in Japan, introduced its first-generation AI accelerator, SAKURA-I, in 2022, claiming over 10× performance-per-watt advantage over competing AI inference solutions based on GPUs for real-time edge applications. It also announced the open-source release of its MERA compiler software framework.
The company has leveraged feedback from customers using its first-generation silicon to deliver an updated architecture that offers future-proofing in terms of activation functions and support for a changing AI model landscape, while adding new features, such as sparse computation, advanced power management, mixed-precision support and a new reshaper engine. It also offers better performance per watt and higher memory capacity in multiple form factors.
SAKURA-II edge AI accelerator (Source: EdgeCortix Inc.)
The SAKURA-II accelerator or co-processor is a compelling solution, with a performance of 60 trillion operations per second (TOPS) with 8 W of typical power consumption, mixed-precision support and built-in memory-compression capabilities, said Sakyasingha Dasgupta, CEO and founder of EdgeCortix. Whether running traditional AI models or the latest GenAI solutions at the edge, it is one of the most flexible and power-efficient accelerators, he added.
SAKURA-II can handle complex tasks, such as large language models (LLMs), large vision models (LVMs) and multimodal transformer-based applications in the manufacturing, Industry 4.0, security, robotics, aerospace and telecommunications industries. It can manage multi-billion parameter models, such as Llama 2, Stable Diffusion, DETR and ViT, within a typical power envelope of 8 W.
In a nutshell, the SAKURA-II accelerator is optimized for GenAI and real-time data streaming with low latency. It delivers exceptional energy efficiency (touted as more than 2× the AI compute utilization of other solutions), higher DRAM capacity—up to 32 GB—to handle complex vision and GenAI workloads, and up to 4× more DRAM bandwidth than other AI accelerators and advanced power management for ultra-high-efficiency modes. It also adds sparse computation to reduce the memory bandwidth, a new integrated tensor reshaper engine to minimize the host CPU load, arbitrary activation function and software-enabled mixed precision for near-F32 accuracy.
EdgeCortix’s solutions aim to reduce the cost, power and time of data transfer by moving more AI processing to the site of data creation. The edge AI accelerator platform addresses two big challenges due to “the explosive growth of the latest-generation AI models,” Dasgupta said. The first challenge is the rising computational demand due to these “exponentially growing models” and the resulting rise in hardware costs, he said.
The cost of deploying a solution or the cost of operation, whether it is in a smart city, robotics or the aerospace industry in an edge environment, is critically important, he added.
The second challenge is how to build more power-efficient systems. “Unfortunately, the majority of today’s AI models are a lot more power-hungry in terms of both electricity consumption as well as carbon emissions,” he said. “So how do we build systems with a software and hardware combination that is a lot more energy-efficient, especially for an edge environment that is constrained by power, weight and size? That really drives who we are as a company.”
The company’s core mission, Dasgupta said, is to deliver a solution that brings near-cloud-level performance to the edge environment, while also delivering orders of magnitude better power efficiency.
Performance per watt is a key factor for customers, Dasgupta added, especially in an edge environment, and within that, real-time processing becomes a critical factor.
Data at the edge
Looking at the data center versus edge landscape, most of the data consumed is being generated or processed, especially enterprise data, at the edge and that will continue into the future, Dasgupta said.
According to IDC, 74 zettabytes of data will be generated at the edge by 2025. Moving this enormous amount of data continuously from the edge to the cloud is expensive both in terms of power and time, he adds. “The fundamental tenet of AI is how we bring computation and intelligence to the seat of data creation.”
Dasgupta said EdgeCortix has achieved this by pairing its design ethos of software first with power-efficient hardware to reduce the cost of data transfer.
The latest product caters to the GenAI landscape as well as across multi-billion parameter LLMs and vision models within the constraint of low-power requirements at the edge across applications, Dasgupta said. The latest low-power GenAI solution targets a range of verticals, from smart cities, smart retail and telecom to robotics, the factory floor, autonomous vehicles and even military/aerospace.
The platform
The software-driven unified platform is comprised of the SAKURA AI accelerator, MERA compiler and framework, DNA technology and the AI accelerator modules and boards. It supports both the latest GenAI and convolutional models.
Designed for flexibility and power efficiency, SAKURA-II offers high memory bandwidth, high accuracy and compact form factors. By leveraging EdgeCortix’s latest-generation runtime reconfigurable neural processing engine, DNA-II, SAKURA-II can provide high power efficiency and real-time processing capabilities while simultaneously executing multiple deep neural network models with low latency.
Dasgupta said it is difficult to put a number to latency because it depends on the AI models and applications. Most applications, depending on the larger models, will be below 10 ms, and in some cases, it could be in the sub-millisecond range.
The SAKURA-II hardware and software platform delivers flexibility, scalability and power efficiency. (Source: EdgeCortix Inc.)
The SAKURA-II platform, with the company’s MERA software suite, features a heterogeneous compiler platform, advanced quantization and model calibration capabilities. This software suite includes native support for development frameworks, such as PyTorch, TensorFlow Lite and ONNX. MERA’s flexible host-to-accelerator unified runtime can scale across single, multi-chip and multi-card systems at the edge. This significantly streamlines AI inferencing and shortens deployment times.
In addition, the integration with the MERA Model Zoo offers a seamless interface to Hugging Face Optimum and gives users access to an extensive range of the latest transformer models. This ensures a smooth transition from training to edge inference.
“One of the exciting elements is a new MERA model library that creates a direct interface with Hugging Face where our customers can bring a large number of current-generation transformer models without having to worry about portability,” Dasgupta said.
MERA Software supports diverse neural networks, from convolutions to the latest GenAI models. (Source: EdgeCortix Inc.)
By building a software-first—Software 2.0—architecture with new innovations, Dasgupta said the company has been able to get an order or two orders of magnitude increase in peak performance per watt compared with general-purpose systems, using CPUs and GPUs, depending on the application.
“We are able to preserve high accuracy [99% of FP32] for applications within those constrained environments of the edge; we’re able to deliver a lot more performance per watt, so better efficiency, and much higher speed in terms of the latency-sensitive and real-time critical applications, especially driven by multimodal-type applications with the latest generative AI models,” Dasgupta said. “And finally, a lower cost of operations in terms of even performance per dollar, preserving a significant advantage compared with other competing solutions.”
This drives edge AI accelerator requirements, which has gone into the design of the latest SAKURA product, Dasgupta said.
More details on SAKURA-II
SAKURA-II can deliver up to 60 TOPS of 8-bit integer, or INT8, performance and 30 trillion 16-bit brain floating-point operations per second (TFLOPS), while also supporting built-in mixed precision for handling next-generation AI tasks. The higher DRAM bandwidth targets the demand for higher performance for LLMs and LVMs.
Also new for the SAKURA-II are the advanced power management features, including on-chip power gating and power management capabilities for ultra-high-efficiency modes and a dedicated tensor reshaper engine to manage complex data permutations on-chip and minimize host CPU load for efficient data handling.
Some of the key architecture innovations include sparse computation that natively supports memory footprint optimization to reduce the amount of memory required to move large amounts of models, especially multi-billion-parameter LLMs, Dasgupta said, which has a big impact on performance and power consumption.
The built-in advanced power management mechanisms can switch off different parts of the device while an application is running in a way that can trade off power versus performance, Dasgupta said, which enhances the performance per watt in applications or models that do not require all 60 TOPS.
“We also added a dedicated IP in the form of a new reshaper engine on the hardware itself, and this has been designed to handle large tensor operations,” Dasgupta said. This dedicated engine on-chip improves power as well as reduces latency even further, he added.
Dasgupta said the accelerator architecture is the key building block of the SAKURA-II’s performance. “We have much higher utilization of our transistors on the semiconductor as compared with some of our competitors, especially from a GPU perspective. It is typically somewhere on average 2× better utilization, and that gives us much better performance per watt.”
SAKURA-II also adds support for arbitrary activation functions on the hardware, which Dasgupta calls a future-proofing mechanism, so as new types of arbitrary activation functions come in, they can be extended to the user without having to change the hardware.
It also offers mixed-precision support on the software and hardware to trade off between performance and accuracy. Running some parts of a model at a higher precision and others at a reduced precision, depending on the application, becomes important in multimodal cases, Dasgupta said.
SAKURA-II is available in several form factors to meet different customer requirements. These include a standalone device in a 19 × 19-mm BGA package, a M.2 module with a single device and PCIe cards with up to four devices. The M.2 module offers 8 or 16 GB of DRAM and is designed for space-constrained applications, while the single (16-GB DRAM) and dual (32-GB DRAM) PCIe cards target edge server applications.
SAKURA-II addresses space-constrained environments with the M.2 module form factor, supporting both x86 or Arm systems, and delivers performance by supporting multi-billion-parameter models as well as traditional vision models, Dasgupta said.
“The latest-generation product supports a very powerful compiler and software stack that can marry our co-processor with existing x86 or Arm systems working across different types of heterogeneous landscape in the edge space.”
SAKURA-II M.2 module (Source: EdgeCortix Inc.)
The unified platform also delivers a high amount of compute, up to 240 TOPS, in a single PCIe card with four devices with under 50 W of power consumption.
Dasgupta said power has been maintained at previous levels with the SAKURA-II, so customers are getting a much higher performance per watt. The power consumption is typically about 8 W for the most complex AI models and even less for some applications, he said.
SAKURA-II will be available as a standalone device, with two M.2 modules with different DRAM capacities (8 GB and 16 GB), and single- and dual-device low-profile PCIe cards. Customers can reserve M.2 modules and PCIe cards for delivery in the second half of 2024. The accelerators, M.2 modules and PCIe cards can be pre-ordered.
Competition? They are moving away from the cloud?
Edge AI accelerator delivers 60 TOPS at minimal power — Electronic Products
EdgeCortix’s SAKURA-II energy-efficient edge AI accelerator handles the latest GenAI tasks at the edge, including LLMs and LVMs.apple.news
Edge AI accelerator delivers 60 TOPS at minimal power
Jul 3, 2024 | 7:31 AM
Startup EdgeCortix Inc. recently launched its next-generation SAKURA-II edge AI accelerator, designed for processing generative AI (GenAI) workloads at the edge with minimal power consumption and low latency. This platform is paired with the company’s second-generation dynamic neural accelerator (DNA) architecture to address the most challenging GenAI tasks in the industry.
EdgeCortix, headquartered in Japan, introduced its first-generation AI accelerator, SAKURA-I, in 2022, claiming over 10× performance-per-watt advantage over competing AI inference solutions based on GPUs for real-time edge applications. It also announced the open-source release of its MERA compiler software framework.
The company has leveraged feedback from customers using its first-generation silicon to deliver an updated architecture that offers future-proofing in terms of activation functions and support for a changing AI model landscape, while adding new features, such as sparse computation, advanced power management, mixed-precision support and a new reshaper engine. It also offers better performance per watt and higher memory capacity in multiple form factors.
SAKURA-II edge AI accelerator (Source: EdgeCortix Inc.)
The SAKURA-II accelerator or co-processor is a compelling solution, with a performance of 60 trillion operations per second (TOPS) with 8 W of typical power consumption, mixed-precision support and built-in memory-compression capabilities, said Sakyasingha Dasgupta, CEO and founder of EdgeCortix. Whether running traditional AI models or the latest GenAI solutions at the edge, it is one of the most flexible and power-efficient accelerators, he added.
SAKURA-II can handle complex tasks, such as large language models (LLMs), large vision models (LVMs) and multimodal transformer-based applications in the manufacturing, Industry 4.0, security, robotics, aerospace and telecommunications industries. It can manage multi-billion parameter models, such as Llama 2, Stable Diffusion, DETR and ViT, within a typical power envelope of 8 W.
In a nutshell, the SAKURA-II accelerator is optimized for GenAI and real-time data streaming with low latency. It delivers exceptional energy efficiency (touted as more than 2× the AI compute utilization of other solutions), higher DRAM capacity—up to 32 GB—to handle complex vision and GenAI workloads, and up to 4× more DRAM bandwidth than other AI accelerators and advanced power management for ultra-high-efficiency modes. It also adds sparse computation to reduce the memory bandwidth, a new integrated tensor reshaper engine to minimize the host CPU load, arbitrary activation function and software-enabled mixed precision for near-F32 accuracy.
EdgeCortix’s solutions aim to reduce the cost, power and time of data transfer by moving more AI processing to the site of data creation. The edge AI accelerator platform addresses two big challenges due to “the explosive growth of the latest-generation AI models,” Dasgupta said. The first challenge is the rising computational demand due to these “exponentially growing models” and the resulting rise in hardware costs, he said.
The cost of deploying a solution or the cost of operation, whether it is in a smart city, robotics or the aerospace industry in an edge environment, is critically important, he added.
The second challenge is how to build more power-efficient systems. “Unfortunately, the majority of today’s AI models are a lot more power-hungry in terms of both electricity consumption as well as carbon emissions,” he said. “So how do we build systems with a software and hardware combination that is a lot more energy-efficient, especially for an edge environment that is constrained by power, weight and size? That really drives who we are as a company.”
The company’s core mission, Dasgupta said, is to deliver a solution that brings near-cloud-level performance to the edge environment, while also delivering orders of magnitude better power efficiency.
Performance per watt is a key factor for customers, Dasgupta added, especially in an edge environment, and within that, real-time processing becomes a critical factor.
Data at the edge
Looking at the data center versus edge landscape, most of the data consumed is being generated or processed, especially enterprise data, at the edge and that will continue into the future, Dasgupta said.
According to IDC, 74 zettabytes of data will be generated at the edge by 2025. Moving this enormous amount of data continuously from the edge to the cloud is expensive both in terms of power and time, he adds. “The fundamental tenet of AI is how we bring computation and intelligence to the seat of data creation.”
Dasgupta said EdgeCortix has achieved this by pairing its design ethos of software first with power-efficient hardware to reduce the cost of data transfer.
The latest product caters to the GenAI landscape as well as across multi-billion parameter LLMs and vision models within the constraint of low-power requirements at the edge across applications, Dasgupta said. The latest low-power GenAI solution targets a range of verticals, from smart cities, smart retail and telecom to robotics, the factory floor, autonomous vehicles and even military/aerospace.
The platform
The software-driven unified platform is comprised of the SAKURA AI accelerator, MERA compiler and framework, DNA technology and the AI accelerator modules and boards. It supports both the latest GenAI and convolutional models.
Designed for flexibility and power efficiency, SAKURA-II offers high memory bandwidth, high accuracy and compact form factors. By leveraging EdgeCortix’s latest-generation runtime reconfigurable neural processing engine, DNA-II, SAKURA-II can provide high power efficiency and real-time processing capabilities while simultaneously executing multiple deep neural network models with low latency.
Dasgupta said it is difficult to put a number to latency because it depends on the AI models and applications. Most applications, depending on the larger models, will be below 10 ms, and in some cases, it could be in the sub-millisecond range.
The SAKURA-II hardware and software platform delivers flexibility, scalability and power efficiency. (Source: EdgeCortix Inc.)
The SAKURA-II platform, with the company’s MERA software suite, features a heterogeneous compiler platform, advanced quantization and model calibration capabilities. This software suite includes native support for development frameworks, such as PyTorch, TensorFlow Lite and ONNX. MERA’s flexible host-to-accelerator unified runtime can scale across single, multi-chip and multi-card systems at the edge. This significantly streamlines AI inferencing and shortens deployment times.
In addition, the integration with the MERA Model Zoo offers a seamless interface to Hugging Face Optimum and gives users access to an extensive range of the latest transformer models. This ensures a smooth transition from training to edge inference.
“One of the exciting elements is a new MERA model library that creates a direct interface with Hugging Face where our customers can bring a large number of current-generation transformer models without having to worry about portability,” Dasgupta said.
MERA Software supports diverse neural networks, from convolutions to the latest GenAI models. (Source: EdgeCortix Inc.)
By building a software-first—Software 2.0—architecture with new innovations, Dasgupta said the company has been able to get an order or two orders of magnitude increase in peak performance per watt compared with general-purpose systems, using CPUs and GPUs, depending on the application.
“We are able to preserve high accuracy [99% of FP32] for applications within those constrained environments of the edge; we’re able to deliver a lot more performance per watt, so better efficiency, and much higher speed in terms of the latency-sensitive and real-time critical applications, especially driven by multimodal-type applications with the latest generative AI models,” Dasgupta said. “And finally, a lower cost of operations in terms of even performance per dollar, preserving a significant advantage compared with other competing solutions.”
This drives edge AI accelerator requirements, which has gone into the design of the latest SAKURA product, Dasgupta said.
More details on SAKURA-II
SAKURA-II can deliver up to 60 TOPS of 8-bit integer, or INT8, performance and 30 trillion 16-bit brain floating-point operations per second (TFLOPS), while also supporting built-in mixed precision for handling next-generation AI tasks. The higher DRAM bandwidth targets the demand for higher performance for LLMs and LVMs.
Also new for the SAKURA-II are the advanced power management features, including on-chip power gating and power management capabilities for ultra-high-efficiency modes and a dedicated tensor reshaper engine to manage complex data permutations on-chip and minimize host CPU load for efficient data handling.
Some of the key architecture innovations include sparse computation that natively supports memory footprint optimization to reduce the amount of memory required to move large amounts of models, especially multi-billion-parameter LLMs, Dasgupta said, which has a big impact on performance and power consumption.
The built-in advanced power management mechanisms can switch off different parts of the device while an application is running in a way that can trade off power versus performance, Dasgupta said, which enhances the performance per watt in applications or models that do not require all 60 TOPS.
“We also added a dedicated IP in the form of a new reshaper engine on the hardware itself, and this has been designed to handle large tensor operations,” Dasgupta said. This dedicated engine on-chip improves power as well as reduces latency even further, he added.
Dasgupta said the accelerator architecture is the key building block of the SAKURA-II’s performance. “We have much higher utilization of our transistors on the semiconductor as compared with some of our competitors, especially from a GPU perspective. It is typically somewhere on average 2× better utilization, and that gives us much better performance per watt.”
SAKURA-II also adds support for arbitrary activation functions on the hardware, which Dasgupta calls a future-proofing mechanism, so as new types of arbitrary activation functions come in, they can be extended to the user without having to change the hardware.
It also offers mixed-precision support on the software and hardware to trade off between performance and accuracy. Running some parts of a model at a higher precision and others at a reduced precision, depending on the application, becomes important in multimodal cases, Dasgupta said.
SAKURA-II is available in several form factors to meet different customer requirements. These include a standalone device in a 19 × 19-mm BGA package, a M.2 module with a single device and PCIe cards with up to four devices. The M.2 module offers 8 or 16 GB of DRAM and is designed for space-constrained applications, while the single (16-GB DRAM) and dual (32-GB DRAM) PCIe cards target edge server applications.
SAKURA-II addresses space-constrained environments with the M.2 module form factor, supporting both x86 or Arm systems, and delivers performance by supporting multi-billion-parameter models as well as traditional vision models, Dasgupta said.
“The latest-generation product supports a very powerful compiler and software stack that can marry our co-processor with existing x86 or Arm systems working across different types of heterogeneous landscape in the edge space.”
SAKURA-II M.2 module (Source: EdgeCortix Inc.)
The unified platform also delivers a high amount of compute, up to 240 TOPS, in a single PCIe card with four devices with under 50 W of power consumption.
Dasgupta said power has been maintained at previous levels with the SAKURA-II, so customers are getting a much higher performance per watt. The power consumption is typically about 8 W for the most complex AI models and even less for some applications, he said.
SAKURA-II will be available as a standalone device, with two M.2 modules with different DRAM capacities (8 GB and 16 GB), and single- and dual-device low-profile PCIe cards. Customers can reserve M.2 modules and PCIe cards for delivery in the second half of 2024. The accelerators, M.2 modules and PCIe cards can be pre-ordered.
" According to IDC, 74 zettabytes of data will be generated at the edge by 2025. Moving this enormous amount of data continuously from the edge to the cloud is expensive both in terms of power and time, he adds. “The fundamental tenet of AI is how we bring computation and intelligence to the seat of data creation.”"
It is a fact that I did not find this paper myself, but it contains tables of an extensive survey of AI chips, conducted by AFRL in conjunction with Dayton Uni.
Table 2 compares 100 AI chips, so it is useful as a compressed, if not 100% accurate, guide to the competition.
https://www.preprints.org/manuscript/202407.0025/download/final_file
...
BrainChip introduced the Akida line of spiking processors. The AKD1000 has 80 NPUs, 3 pJ/synaptic operation, and around 2 W of power consumption [147]. Each NPU consists of eight neural processing engines that run simultaneously and control convolution, pooling, and activation (ReLu) operations [148]. Convolution is normally carried out in INT8 precision, but it can be programmed for INT 1, 2, 3 or 4 precisions while sacrificing 1-3% accuracy. BrainChip has announced future releases of smaller and larger Akida processors under the AKD500, AKD1500, and AKD2000 labels [148]. A trained DNN network can be converted to SNN by using the CNN2SNN tool in the Meta-TF framework for loading a model into an Akida processor. This processor also has on-chip training capability, thus allowing the training of SNNs from scratch by using the Meta-TF framework [146].
...
Several of the edge processors offer on-chip retraining in real-time. This enables retraining of networks without having to send sensitive data to the cloud, thus increasing security and privacy. Intel’s Loihi 2 and Brainchip’s Akida processor can be retrained on local data for personalized applications and faster response rates.
Table 2 compares 100 AI chips, so it is useful as a compressed, if not 100% accurate, guide to the competition.
https://www.preprints.org/manuscript/202407.0025/download/final_file
...
BrainChip introduced the Akida line of spiking processors. The AKD1000 has 80 NPUs, 3 pJ/synaptic operation, and around 2 W of power consumption [147]. Each NPU consists of eight neural processing engines that run simultaneously and control convolution, pooling, and activation (ReLu) operations [148]. Convolution is normally carried out in INT8 precision, but it can be programmed for INT 1, 2, 3 or 4 precisions while sacrificing 1-3% accuracy. BrainChip has announced future releases of smaller and larger Akida processors under the AKD500, AKD1500, and AKD2000 labels [148]. A trained DNN network can be converted to SNN by using the CNN2SNN tool in the Meta-TF framework for loading a model into an Akida processor. This processor also has on-chip training capability, thus allowing the training of SNNs from scratch by using the Meta-TF framework [146].
...
Several of the edge processors offer on-chip retraining in real-time. This enables retraining of networks without having to send sensitive data to the cloud, thus increasing security and privacy. Intel’s Loihi 2 and Brainchip’s Akida processor can be retrained on local data for personalized applications and faster response rates.
Space Cadet
Regular
Tata Elxsi have been talking up health care. Found this interesting. Noo mention of Akida but with NDA's you wouldn't expect to.
Show where they are heading with Neuromorphic and Healthcare. There are so many avenues for us.
https://ai.tataelxsi.com/healthcare...or-disease-detection-diagnosis-and-prognosis/
Apologies if already posted.
SC
Show where they are heading with Neuromorphic and Healthcare. There are so many avenues for us.
https://ai.tataelxsi.com/healthcare...or-disease-detection-diagnosis-and-prognosis/
Apologies if already posted.
SC
Similar threads
