The cloud and software specialist has launched a number of cloud-based applications within its Oracle Cloud Infrastructure portfolio that make certain workloads for High Performance Computing (HPC) cheaper.
According to the cloud and software vendor, the solutions now presented are particularly suitable for companies that have not yet made the choice to perform their heavy HPC tasks, such as artificial intelligence or heavy technical simulations, in a cloud environment.
According to Oracle, the main reason for this was that they had not yet been able to carry out the tasks in a cloud environment due to the absence of cloud environments that offer high performance in terms of computing power and at the same time were not too expensive to carry out these tasks. As a result, they continued to perform these tasks on expensive on-premise equipment.
Part of comprehensive computer network
Oracle’s cloud-based solutions, which have now come onto the market, must make the latter a thing of the past. HPC workloads are performed on bare-metal computer instances in a cloud environment without compromising on performance. At the same time, these cloud solutions also offer more flexibility and lower costs.
In addition, the solutions presented are part of the new “Clustered Network” of the cloud and software specialist. This network gives end users access to a so-called RDMA network with low latency and high bandwidth. In this way, end users can benefit even more from Oracle’s possibilities for extremely high computing power for the heaviest applications.
The new solutions are based on high-frequency and scalable Intel Xeon processors and Mellanox’s high-performance network interface controllers. This configuration completes the comprehensive set of Oracle infrastructure solutions for the full range of CPU and Nvidia GPU-based HPC workloads. Oracle Cloud Infrastructure now offers a complete range of HPC workloads, including the only bare-metal Infrastructure-as-a-Service (IaaS) offering on the market with RDMA.
In addition, these possibilities will also be extended to Oracle’s recently announced support for the Nvidia HGX-2 platform on Oracle Cloud Infrastructure. Customers can take advantage of clustered networks as part of the next generation Nvidia Tensor Core GPUs, allowing them to open up large-scale GPU workloads such as artificial intelligence and deep learning.
Paying via pay-per-use
Oracle’s cloud-based solutions are available in the United States and Europe via a pay-per-use model. The cost is $0.0075 per kernuur. The software and cloud specialist says it’s 49 percent cheaper than comparable services offered by other cloud providers.This news article was automatically translated from Dutch to give Techzine.eu a head start. All news articles after September 1, 2019 are written in native English and NOT translated. All our background stories are written in native English as well. For more information read our launch article.