DevCloud program offers users easy access to develop, test, and run their HPC and AI applications.
The QCT DevCloud, in collaboration with Intel, is a comprehensive HPC, AI and data analytics environment for users to experience QCT solutions and test their workloads remotely on QCT innovative platforms. The QCT DevCloud program facilitates the exchange of user workload experiences and QCT’s infrastructure expertise. The program includes support for QCT precompiled workloads and Intel development toolkits across various QCT hardware platforms for various HPC and AI workloads.
Fig. 1. HPC and AI workloads
As a cloud-based heterogeneous platform, the QCT DevCloud provides the latest QCT QuantaGrid, QuantaMesh, and QuantaVault family of hardware to meet diverse HPC & AI demands to perform complex computations with access to various development tools and faster computing remotely. Supporting cloud-native & bare-metal environments with resource and job management tools, users also gain access to a powerful robust development environment to speed up their testing process with flexibility. Powered by the Intel® Xeon® Scalable processors, users also gain the built-in features of Intel® Software Guard Extensions (Intel® SGX) and Intel® Advanced Matrix Extensions (Intel® AMX) which enhance the security level and accelerate the development of HPC and AI workloads. Furthermore, Intel® oneAPI and its toolkits offer a powerful software stack, increasing productivity for their development journey.
Fig. 2. QCT DevCloud Benefits for Developers
Through the QCT DevCloud, we compared the 3rd Gen and 4th Gen Intel Xeon Scalable platforms to understand the performance acceleration and scalability on the 4th Gen platform compared to the 3rd Gen platform as the number of nodes was increased using VASP (Vienna Ab initio Simulation Package) and Quantum Espresso (QE). Additionally, credit card fraud, which poses a significant threat and is a compute-intensive operation that is greatly needed as the number of incidents continues to increase. Machine learning (ML) can play a vital role in detecting fraudulent transactions and considering the scale at which these transactions happen, an ML approach is a commonly implemented solution. Through the QCT DevCloud program, we are able to accelerate the cluster analysis and inference process with less compute intensive resources over the stock packages using the Intel® AI Analytics Toolkit. It helps to greatly reduce the total cost of ownership and the time it takes to detect a potential fraud instance. Data scientists can also benefit from accelerating clustering during training to achieve more accurate models, which further enhances fraud detection. These additional performance gains enabled by the Intel AI Kit creates a more efficient fraudulent detection system without the latency overhead.
QCT DevCloud offers features and benefits that make it a valuable platform to researchers and developers based on the acceleration improvements. With access to Intel’s cutting-edge technologies and toolkits, customizable environment, and QCT infrastructure expertise, users can work with comprehensive HPC/AI environments to develop, test, and optimize their applications for many use cases. Additionally, leveraging the full potential of Intel’s toolkits and capabilities, it’s clear that this platform can play a vital role in driving innovation and progress in HPC and AI.
Register here to join the QCT DevCloud Program