GPU Elastic Physical Compute provides services including hardware (GPU, CPU, memory, and hard disk) and network. The price of the instance hardware is in the table below.
Supports: GPU Physical Instance Standard and GPU Physical Instance Standard Cluster.
|GPU Physical Instance Standard||P3E||
It provides a total of 48TFlops single-precision floating-point computing power and 96GB of memory, which can be used for large-scale neural network training and online reasoning.
|GPU Physical Instance Standard Cluster||P3E-R||
It supports 100Gbps lossless non-converged cluster networking with 256 nodes, and can be able to support up to 1024 P40 GPU parallel computing as well. This model uses RoCE V2's RDMA network, with low latency and high bandwidth, which is suitable for extreme large-scale neural network training and scientific computing.
|Type||GPU (Tesla P40)||GDDR5 (GB)||vCPU (Core)
||Local SSD||100Gbps RDMA Network||
|P3E-R (Standard Cluster)