A new supercomputer has arrived in the world. America’s AMD-powered Frontier supercomputer is now the fastest in the world, overtaking Japan’s Fugaku as the fastest supercomputer by a margin, according to the latest edition of the top 500 list of the world’s fastest supercomputers.
According to the list of the Top 500 Supercomputers, the Frontier has a performance of 1.102 exaflops, more than twice the power of Japan’s Fugaku. In fact, Frontier is faster than the next seven supercomputers on the list combined. During testing, the Frontier was capable of delivering 1.69 ExaFlops in maximum performance. One ExaFlop is equal to one quintillion floating point operations per second. Earlier, Fugaku had put out 442 petaflops of power. Petaflop means one quadrillion floating-point operations per second.
Frontier also ranks as the fastest AI system in the world, delivering 6.88 ExaFlops of mixed-precision performance. This equates to 68 million instructions per second for every 86 billion neurons in the brand, giving you an idea of the amount of computing power the fastest. super computer holds. Frontier is also number one on the Green 500 list, which means it is also the most power efficient supercomputer in the world.
The Frontier supercomputer is built by HPE and installed at the US Department of Energy’s (DOE) Oak Ridge National Laboratory in Tennessee. The system has 9,408 compute nodes, each with a 64-core amd The CPU is paired with 512GB of DDR4 RAM and four AMD Radeon Instinct MI250X GPUs. The 74 HPE Cray EX cabinet consists of scattered nodes, each weighing over 3,600 kg. In addition, the CPU and GPU are tied together using the Ethernet-based HPE Cray Slingshot-11 Networking Fabric. The entire system uses water cooling to keep the heat under control, with 6,000 gallons of water being moved through the system by 350-horsepower pumps—these pumps can fill an Olympic-sized swimming pool in just 30 minutes !
Read also: Apple may unveil RealityOS for mixed reality headsets during WWDC 2022: Report
To assemble the Frontier, engineers had to source 60 million parts with 685 different part numbers to build the system. The chip went missing during manufacture, affecting 167 of those part numbers.
Chip shortages also slowed manufacturing as it affected 167 of those parts, so ORNL found itself two million parts short. AMD also ran into issues as it suffered a 200 part number reduction for the MI 15 GPU.
In terms of power consumption, the system peaks at 29 megawatts, but Frontier’s mechanical plant can cool up to 40 megawatts of computational power, the equivalent of 30,000 US homes. The plant can be scaled up to 70 MW, leaving room for future growth.
This supercomputer gave AMD a major edge over rival Intel, whose summit was the world’s most powerful supercomputer before Japan’s Fugaku. IntelHowever, Aurora is working to build a supercomputer that is said to come with up to 2 exaflops of performance.
read all breaking news , today’s fresh news And IPL 2022 Live Updates Here.