Taiwanese Semiconductor Manufacturing Company has announced that it is now using AMD’s EPYC CPUs to manufacture brand new and next-generation chips. Utilizing AMD’s EPYC line of chips will offer a clear advantage over previous implementations for the world’s largest chip manufacturer.
TSMC earlier stated that each automation machine needs to have access to one x86 server to control the operation speed, provision of water, electricity & gas (power input). And with the help of AMD’s EPYC CPUs TSMC is designing the current and next-generation chips for consumers and HPC powerhouses.
“We first introduced AMD EPYC processors into the general workload,” explains Simon Wang, Director of Infrastructure and Communication Services Division at TSMC, “and they are being deployed with our research and development team.” TSMC was looking for a server solution optimized to implement hyper-converged infrastructure (HCI) design where all three components — compute, storage, and networking — could be provided by the same underlying hardware.
Although TSMC is the company that manufactures AMD’s technology-leading 7nm products, which includes the groundbreaking AMD EPYC, Ryzen, and Ryzen Threadripper processors, as well as the AMD Radeon, Radeon Pro, and AMD Instinct GPUs, it didn’t mean that AMD EPYC processors were automatically going to be the preferred selection.
Until recently, TSMC was relying on the HPE DL32G 510 platform which comprises a 2nd Gen AMD EPYC Rome 7702P CPU featuring 64 cores and a total of 128 threads. It clocks at a 2 GHz base frequency and 3.35 GHz boost clock.
“For automation with the machinery inside our fab, each machine needs to have one x86 server to control the operation speed and provision of water, electricity, and gas, or power consumption,” adds Wang.
These machines are very costly. They might cost billions of dollars, but the servers that control them are much cheaper. I need to make sure that we have high availability in case one rack is down, then we can use another rack to support the machine. With a standard building block, I can generate about 1,000 virtual machines, which can control 1,000 fab tools in our cleanroom.” This will mean a huge cost saving without sacrificing failover redundancy or reliability.
According to the chip manufacturer, using AMD EPYC chips its system can generate about 1000 virtual machines with a standard block which is very impressive. This marks a big win for AMD’s EPYC CPUs in powering the world’s leading semiconductor manufacturing company.