9 May 2017 San Jose - Inspur, an expert in intelligent computing and ranked by Gartner as a top 5 server manufacturer worldwide, unveiled its new AI deep learning training cluster management software, AIStation at GTC 2017. Inspur AIStation is a complete AI deep learning cluster management software enabling simple and flexible deep learning, and makes it easy to train and combine popular model types across multiple GPUs and servers. It's also designed with speed, ease of control and efficient resource management in mind, allowing for a more flexible way to organize computation and puts the power of deep learning into the hands of engineers and data scientists.
The AIStation accelerates widely-used deep learning frameworks such as Caffe, TensorFlow and CNTK, as well as popular deep learning models, such as GoogLeNet, VGG and Resnet. It supports real-time monitoring and visualization training process, records the value of the loss function on data, training or testing errors, and supports the dynamic allocation of GPU resources to improve the efficiency. The AIStation provides a unique set of tools allowing you to not only effortlessly apply deep learning to your problems but to deploy the solutions into productions with literally one click. Overall, the AIStation is a complete cluster management software with tools for control, analysis and productivity.
Inspur is a major server vendor that offers FPGA based deep learning accelerators. Currently, Inspur is the main AI GPU server provider for three out of the top seven global internet companies.