The AI-based applications are rapidly expanding in the market, also it has increased a number of deployments in various embedded fields. On the other hand, the requirement of AI performance also increases furthermore in the image-based AI inference (hereafter, “Vision AI”) due to rapidly evolving of the neural network itself. With this trend, it becomes apparent the heat generation issue by power consumption.
In this article, I will introduce the solution to the biggest challenge in AI inference that is "heat generation", which you have already faced or will face.
As you know better than I do, when the heat generation is increased, you need to make countermeasure to install heat sink and fan, or bigger casing to stop increasing the temperature in the case. But these countermeasures are one of reason your product becomes expensive and bigger.
Also, when your system operates for a long time, thermal throttling* and other heat-oriented problems may occur. As a result, performance of AI inference becomes unstable.
* This is a function when chip becomes high temperature, it reduces the clock frequency of the CPU or AI accelerator.
Based on the above, we defined all necessary conditions for high-loaded endpoint Vision AI are "practical AI performance", "low power consumption", and "stable AI inference performance".
Renesas released RZ/V series to solve these conditions. RZ/V series utilizes our technology and know-hows that supports embedded market for a long time. I will introduce “RZ/V2M” that is the first product of RZ/V series, in terms of AI inference.
1. What is RZ/V2M?
RZ/V2M has DRP-AI as key technology to accelerate the AI inference, and it has various peripheral function. In addition, RZ/V2M achieved less than 4W* power consumption during AI inference by applying various power consumption measures.
* Actual power consumption will depend on the customer's usage conditions.
These are the main features of RZ/V2M.
* Tuned ISP is optimized the parameters by experts for Renesas selected sensors.
RZ/V2M is being adopted such as surveillance camera, industrial cameras, marketing camera, gateway, robot, and medical equipment using AI inference, and these products will be shipped to the market soon.
2. “DRP-AI”: Accelerator for AI inference
The most important feature of DRP-AI is all the processing for AI inference can be performed with DRP-AI alone. DRP-AI consists of the following two blocks, each of which performs AI inference through cooperative operation without CPU.
Processing for except convolution layer which requires various processing using reconfigurable technology. Also, DRP can be treated pre-processing such as resizing before AI inference.
Processing for convolution layer which requires simple calculation with large amount of processing
The benefit of this structure does not use CPU or the others hardware IP. It can be realized the stable AI inference even if the other process is working at the same time. For more technical information of DRP-AI, please refer to “DRP-AI white paper”.
3. Performance of RZ/V2M
In general, the performance of AI inference indicates as “TOPS”. However, this index does not include the factor such as power consumption, so it is difficult to consider the total system using this index. Therefore, Renesas introduces the performance as system point of view. I will introduce AI performance and whole system’s power consumption in this blog.
The “system” includes AI required function such as CMOS input and ISP function, also it includes the power consumption these are RZ/V2M and external devices putting on the evaluation board.
This table shows the result.
|AI performance (fps)||55fps||52fps||46fps|
|Power efficiency||21.3 fps/W||17.2 fps/W||4.7 fps/W|
As this result, RZ/V2M achieved better AI performance with only 30% power consumption from other company’s one. We had been introduced to customers the index of power efficiency (fps/W) is also achieved almost 3.5 times.
The following photo shows appearance and surface temperature of the evaluation board that we measured.
The photo on the left shows the appearance of evaluation board. As you can see, the heatsink is not mounted.
The photo on the right shows the surface temperature after running AI inference for about 30 minutes. Since the surface temperatures of the chip is only 30 degrees Celsius, it is easy to understand RZ/V2M does not require any cooling components. Also, the board made by other company’s reaches around 75°C even with heatsinks to dissipate the heat.
We often hear customers often use GPUs and FPGAs for initial evaluation because it is easy to start evaluating. On the other hand, we also heard from customer “these chips difficult to use these chips for mass production due to its high-power consumption and heat generation”.
In fact, we have received following feedback from customer who have selected RZ/V2M:
- "RZ/V2M is impressive device! It achieves almost same AI performance with only less than 25% power consumption of FPGA."
- "We have deeply impressed with the performance of DRP-AI and have high hopes for equipment AI into our products."
I have confident that you can understand RZ/V2M is the most suitable chip for endpoint Vision AI from this blog.
1. e-AI: endpoint AI proposed by Renesas
2. RZ/V2M: Mid-range product for camera applications with high-performance ISP in addition to DRP-AI
3. RZ/V2L: Entry-level product group equipped with the same DRP-AI as RZ/V2M
4. DRP-AI, DRP-AI White Paper: Technical information for DRP-AI