1.5 million cores PC;Tesla

[sample text]

can someone explain me 2 questions:

What's the usage of NVIDIA Tesla K20X? I know it's a GPU for servers(?),but why people compare a GPU with a CPU? https://www.pcper.com/news/General-Tech/NVIDIA-Launches-Tesla-K20X-Accelerator-Card-Powers-Titan-Supercomputer

They use a GPU,unstead of CPU? Like ppl are using GPU for bitcoin farming,instead of CPU?

 

The second question is:

How a library has 1,500,000 cores computer?This means (for example) that they bought x53,571 Intel Xeon Platinum 8176F(28 cores;56 threads),because 28*53,571 = 1,500,000?Or they are using something so powerful that it's not even a CPU?If they bought x53,571 Intel 8176F,they spend 85,713,600 dollars only for CPU .Ofc maybe they have some discounts for such amount of shopping lol

 

 

[Jihakuz]

They may have 1.5 million nvidia cuda cores, not 1.5 million phsyical computing cores

[Enderman]

GPUs have far more cores than CPUs, which is why they are used for parallel computing workloads.

[AntiTrust]

Quote

NVIDIA claims to have already shipped 30 PFLOPS worth of GPU accelerated computing power. Interestingly, most of that computing power is housed in the recently unveiled Titan supercomputer. This supercomputer contains 18,688 Tesla K20X (Kepler GK110) GPUs and 299,008 16-core AMD Opteron 6274 processors.

 

 not bad for 2012 

 

 

 

 

[Krystian]

Reason for GPU usage is to perform sets of arithmetic operations only - very fast and efficiently. CPU is good at doing multiple tasks, including arithmetic operations ofcourse, but not as fast and and efficient as GPU which is way better than CPU, but is not capable of what CPU is. Hence people use GPU for mining rather than CPU.

[Krystian]

Sorry, double post.