IBM has announced a significant breakthrough in general-purpose quantum computing, as AMD's standard chips have successfully run a key error correction algorithm.
AMD's FPGAs Manage to Run Qubit Error Correction Algorithms, With Up To 10x Higher than Desired Performance
Well, IBM has been one of the leading entities in the race for quantum computers, excelling alongside Google in terms of milestones with the technology. However, the company is more focused on what appears to me to be 'mainstream developments', and their recent announcement appears to be a huge breakthrough. According to a report by Reuters, IBM has reportedly managed to run a quantum error correction algorithm on AMD's FPGAs, achieving a ten times faster implementation than initially anticipated.
Jay Gambetta, director of IBM research, said the work showed that IBM's algorithm not only works in the real world, but can operate on a readily available AMD chip that is not "ridiculously expensive."
- Reuters
Well, for those unaware of what an error-correction algorithm (QEC) actually means, let's discuss it. Quantum computing is all around 'qubits', which are much different from the classical binary bits we are used to. And, qubits are extremely fragile, which means they are affected by environmental factors during computation, such as a tiny vibration. It is here that error correction algorithms come into play. They are used to identify the errors and fix those without destroying the qubit’s state. This is a really fine-line detail of how QEC works, but it is sufficient for the context of this post.
The reason why AMD's FPGA proves to be a sufficient compute platform for QEC algorithms is that it is reconfigurable hardware, which means customized tasks can be performed with very high efficiency. And, in the case of error correction, the presence of a robust feedback loop is a necessity, which requires extremely low latency; hence, AMD's FPGAs come into play. And more importantly, this has also outsourced a portion of classical quantum computing to "off-the-shelf" hardware, preventing the need for custom silicon.
Well, NVIDIA's quantum computing strategy doesn't rely on fine-level chips like FPGAs; rather, the firm has a comprehensive tech stack surrounding it, involving DGX Quantum with CUDA-Q onboard. They are also sufficient for QEC algorithms and would probably yield better performance relative to FPGAs. However, AMD's achievement here isn't just running the QEC algorithms; rather, it is the use of commodity hardware, something that NVIDIA has yet to achieve. One of the reasons for this is that NVIDIA doesn't have an asset onboard similar to AMD's Xilinx.
Quantum computing is a relatively new venture for the industry, emerging at a time when the AI frenzy is at its peak. And, it would be interesting to see how manufacturers like NVIDIA and AMD evolve once quantum computers become the next 'AI infrastructure'
About the author: Muhammad Zuhair is a hardware and technology reporter for Wccftech, specializing in the semiconductor industry and the complex interplay between technology, manufacturing, and geopolitics. His coverage focuses on the corporate strategies and technological roadmaps of industry giants like TSMC, NVIDIA, Samsung, and Intel. Zuhair's expertise lies in deconstructing complex topics such as fabrication nodes (e.g., 2nm process), the economic impact of policies like the CHIPS Act, and the strategic development of AI infrastructure from NVIDIA, AMD and Intel.
Follow Wccftech on Google to get more of our news coverage in your feeds.
