A Successful Case of How Cloud HPC Facilitates Research and Education in University.
Recently, Cloudam HPC has deployed a cloud high-performance computing platform that supports the hybrid cloud model for the Department of Electronics and Information Engineering of The Hong Kong Polytechnic University (hereinafter PolyU). Through this cooperation, the shortage of computing resources and management problems in the “AI and Robotics” group have been subtly solved, the utilization rate of GPU resources has been increased by more than 30%, and the overall teaching and scientific research efficiency has been increased by 45% without any queuing. The cloud-HPC solution has hugely boosted scientific research and teaching efficiency.
Image via The Hong Kong Polytechnic University Official website
PolyU is a key player in high education around the globe. The “AI and Robotics” group is affiliated with the Department of Electronics and Information Engineering of PolyU, focusing on R&D in image/video content analysis and processing, CV technology, machine learning, and deep learning.
Image via The Hong Kong Polytechnic University Official website
In order to support the teaching and research work of intelligent robots, the Department of Electronics and Information Engineering of PolyU has configured 8 deep learning servers with 4 cards for the group in recent years. However, the application scenarios of intelligent robots have greatly expanded, therefore, the demand for GPU and CPU computing power surges. Also, the problem of intelligent robot computing resource scheduling and management has gradually emerged along the way.
Cloudam HPC integrates the group's local server resources and cloud resources. When the group’s local server computing resources cannot meet its own needs, it supports automatic scaling out to the cloud through the cluster scheduling system platform, and seamlessly connects with the resources on the cloud, so that there is no need for job queuing, thereby improving computing efficiency.
Through the SLURM job cluster scheduling system and parallel computing scheduling management module, HPC administrators are able to clearly check the status of each user's resource usage and submission of training tasks, flexibly configure user resource quotas, dynamically set charging models, and easily view reports to prevent excessive occupation and waste of resources.
In order to ensure the security and privacy of information and data in the laboratory, Cloudam HPC adopts five major security protection devices: AES-256 data transmission encryption on data transmission and staging, and two-way TLS, key pair link cluster, and automatic data clearing at regular intervals.
"Cloudam Cloud-HPC platform supports a variety of submission methods such as visual interface, command line, workstation, etc., and has built in various mainstream deep learning frameworks and development components, which can quickly deploy the development environment and improve the experimental efficiency." Said a reasearcher in The Hong Kong Polytechnic University. "Cloudam has created an efficient R&D environment for us which allows us to devote more energy to teaching and scientific research."
The Department of Electronic and Information Engineering of PolyU leverages the Cloudam Cloud HPC platform to solve problems such as insufficient local computing resources, uneven computing power distribution, equipment management and maintenance, etc. It is a successful example of school-enterprise cooperation that will further accelerate the scientific research and innovation of intelligent robots and the education of talents. PolyU will create a benchmark for intelligent robot teaching and research.
About Cloudam HPC
Cloudam HPC is a one-stop HPC platform with 300+ pre-installed to deploy immediately. The system can smartly schedule compute nodes and dynamically schedule the software licenses, optimizing workflow and boosting efficiency for engineers and researchers.
Partnered with AWS, Azure, Google Cloud, Oracle Cloud, etc., Cloudam powers your R&D with massive cloud resources without queuing.
You can submit jobs by intuitive templates, SLURM, and Windows/Linux workstations. Whether you are a beginner or a professional, you can always find it handy to run and manage your job.
There is a $30 Free Trial for every new user. Why not register and boost your R&D NOW?