IBM unveiled AIU chip: 5nm 32 core, 23 billion transistors
Views : 150
Update time : 2022-11-14 10:57:28
In October, IBM unveiled its first Artificial Intelligence Unit (AIU) system-on-chip, an application-specific integrated circuit (ASIC) designed to train and run deep learning models that require large-scale parallel computing faster and more efficiently.
In the past few years, the industry has mainly used CPUs and GPUs to run deep learning models, but as the number of artificial intelligence models is growing exponentially, deep learning models are becoming larger and larger, with billions or even trillions of parameters. The computing power required is also getting higher and higher, and the growth of AI computing power of chips with traditional architectures such as CPUs and GPUs has encountered bottlenecks.
According to IBM, deep learning models have traditionally relied on a combination of CPUs and GPU coprocessors to train and run models. The flexibility and high precision of CPUs are well suited for general-purpose software applications, however, CPUs are at a disadvantage when it comes to training and running deep learning models that require massively parallel AI operations. Both CPUs and GPUs were designed before the deep learning revolution, and now their efficiency growth has lagged behind the exponential growth of deep learning computing power. What the industry really needs is general-purpose chips optimized for matrix and vector multiplication types for deep learning. Based on this, the IBM Research AI Hardware Center has been focusing on developing next-generation chips and artificial intelligence systems for the past five years, hoping to improve the efficiency of artificial hardware intelligence by 2.5 times by 2029 and be able to train and run artificial intelligence models 1,000 times faster than in 2019. The latest AIU chip is the first IBM chip to be customized for modern AI statistics. IBM says AIU can solve complex problems in vector computation and perform data analytics faster than CPU capabilities. The AIU chip is a complete system-on-chip based on an extended version of the proven AI accelerator built into IBM's previous Telum chip (7nm process) and uses a more advanced 5nm process with 32 processing cores and 23 billion transistors. The IBM AIU is also designed to be as easy to use as a graphics card. It can be plugged into any computer or server with a PCIe slot. IBM said: "Deploying AI to sort cats and dogs in photos is a fun academic exercise. But it won't solve the pressing problems we face today. For us to get AI to solve real-world complexities - like predicting the next hurricane Ian, or whether we're heading for a recession - we need enterprise-grade, industrial-grade hardware. Our AIU takes that vision one step closer."
The LDK220 series of low-dropout linear regulators (LDOs) is a high-performance device designed specifically for low-power consumption and high-precision voltage regulation, widely used in scenarios such as consumer electronics, industrial control, and po
Xilinx Spartan®-7 FPGA Family stands as a defining solution in the mid-range FPGA landscape, blending high performance, energy efficiency, and cost-effectiveness to redefine versatility for industrial, IoT, and consumer electronics applications. Built on
The Altera FLEX series was more than a lineup of FPGAs—it was a blueprint for how programmable logic devices could evolve to meet diverse market needs. The FLEX 8000 laid the architectural groundwork, the FLEX 10K redefined functionality with embedded mem
Among TI standout offerings, the LM4765 and LM4766 are dual-channel amplifiers designed to cater to diverse audio needs—from compact setups to high-fidelity systems. While sharing the same product lineage, these chips differ significantly in power output,
