Cloud Native ComputingDevelopersDevOpsNews

DigitalOcean Launches Premium CPU-Optimized Droplets

0

DigitalOcean has announced the launch of Premium CPU-Optimized Droplets, virtual machines built for high throughput and reliable performance, ideal for network and computing-intensive workloads such as streaming media, online gaming, machine learning, and data analytics. According to the company, the mix of newer CPUs, faster network speeds, and NVMe SSDs makes it easy to scale your apps for data-intensive workloads, especially machine learning and AI applications. Premium CPU-Optimized Droplets are a great choice to train and develop powerful data analysis models for your business and customers.

For bandwidth-intensive use-cases, such as streaming and gaming, the benefits of high performance hardware can be maximized only when the processed data is transferred to the endpoint quickly. With five times higher outbound network speeds when compared to their Regular variant, the new Premium CPU-Optimized Droplets make delivering high-quality digital experiences easier than ever.

Premium CPU-Optimized Droplets utilize some of the latest generations of Intel Xeon CPUs, providing consistently high application performance. Based on the company’s internal benchmark tests, Premium CPU-Optimized Droplets deliver up to a 58% increase in single-core performance and a 20% increase in multi-core performance.

NVMe solid-state drives (SSD) use parallelism to deliver faster disk performance than with regular SSDs. Workloads that require a large number of transactions will have much lower latency with NVMe SSDs. File reads via Premium CPU-Optimized Droplets are up to 65% faster and file writes are up to 290% faster, when compared to Regular CPU-Optimized Droplets.

A dedicated CPU Droplet has guaranteed access to the full hyperthread at all times, delivering a consistently high level of performance. This is unlike a shared CPU Droplet, that may get a fraction of CPU cycles when other Droplets sharing the CPU experience a high load.