Gurugram, Mar 25: AceCloud (a brand of RTDS), a leading provider of cloud computing solutions to global organizations across industries, today announced the launch of GPU Spot Instances on its cloud platform vailable across AceCloud regions in India and the United States. This new offering enables customers to access high-performance GPU computing at significantly reduced costs, making advanced AI and machine learning workloads more accessible and economical.
GPU Spot Instances provide customers with a flexible and cost-effective way to leverage GPUs for workloads that can tolerate interruptions. By availing GPUs through Spot Instances, customers can get cost savings of up to 50% compared to standard on-demand GPU pricing, without compromising performance.
GPU Spot Instance prices will be accessible to customers via a live pricing graph on AceCloud’s website. The graph will showcase real-time prices based on various parameters such as time of the day/week, instance type, current demand, etc. Customers can use this tool to monitor pricing trends and make informed decisions about when and how to use GPU Spot Instances to get the best possible value. By leveraging this dynamic pricing model, users can significantly reduce their operational costs, particularly for non-critical AI tasks or those that can be scheduled for off-peak periods.
Speaking about the launch, Mr. Vinay Chhabra, Co-Founder & Managing Director, AceCloud, a brand of Real Time Data Services, said, “As AI and high-performance computing adoption accelerates, organizations are looking for ways to scale GPU-intensive workloads without significantly increasing costs. With the introduction of GPU Spot Instances, we are extending the benefits of flexible GPU pricing, enabling customers to optimize spend while continuing to innovate. This offering is particularly well suited for checkpointing batch and other workloads where flexibility and cost efficiency are key considerations.”
Ideal for AI and machine learning workloads such as model training, fine-tuning, inference benchmarking and experimentation, data analytics and rendering, customers can avail GPU Spot Instances directly through the AceCloud’s console while launching new GPU resources. They also benefit from AceCloud’s round-the-clock technical support and tools designed to monitor usage, manage capacity and minimize the impact of potential interruptions.
In addition to cost savings, GPU Spot Instances provide businesses and startups with greater flexibility to scale their resources based on evolving workload requirements. By combining GPU Spot Instances with on-demand GPU instances, customers can balance performance, availability and cost, enabling them to respond quickly to changing demands without overcommitting resources. This adaptable approach ensures organizations have access to the GPUs they need, while maintaining control over cloud spend.
