We received an inquiry from a customer regarding a testing machine for performing deep learning based on Parallel Computing Toolbox of Matlab.
In addition to Parallel Computing Toolbox, verification is planned to be performed using environments such as Caffe / Theano / TensorFlow / CUDA / CuDNN / Python, and it is assumed that test operation of each tool will be possible and used without any inconvenience. We are requesting proposals with specifications that we can.
Regarding the GPU, I was hoping for either Maxwell generation TITAN X or Tesla K40 which is Matlab's supported GPU at the time of inquiring, but TITAN X is finished, budget for Tesla K40 is 70 yen Since it doesn't fit within the budget, we focused on budget and compatibility, and also proposed the Maxwell generation Quadro M5000.
【Main Specifications】
| CPU | Xeon E5-1650v4 (3.6GHz 6 cores) |
| memory | 32GB ECC REG |
| storage | HDD 4TB |
| network | GigabitLAN x2 |
| Video | NVIDIA Quadro M5000 |
| Housing + power supply | Middle tower case + 850W |
| OS | Ubuntu 14.04 |
| ■ Click here for details and inquiries about this PC case Deep Learning test machine by Matlab (budget 70) |
In addition to this, Tegsys introduces the proposal results of PC for Deep Leaning.
Deep Learning Machine Proposal Case List
Please feel free to contact us as we will make an optimal proposal according to the customer's request.
