Alibaba Cloud At its Beijing Summit, Alibaba Cloud announce PolarDB, cloud-native relational database service designed for enterprise-grade database applications. PolarDB is compatible with Oracle, MySQL and PostgreSQL. Also at the summit, Alibaba Cloud announced its release of X-dragon Super-Computing Cluster instance SCC-GN6: a super-computing bare-metal server instance with the capability to enhance cluster performance by more than 100%. The instance offers “50 Gbps of RDMA networking throughput, 96 custom Intel Skylake vCPUs, 8 NVIDIA V100 Tensor Core GPUs, providing a total of 1000 TFlops to computational throughput, and high-performance parallel file system CPFS with 1TB/s R/W throughput.” Amazon Web Services (AWS) Last week, AWS Chief Evangelist Jeff Barr gave a glimpse at AWS’s newest GPU-equipped instance, the G4 using NVIDIA T4 GPU. G4 is designed for machine learning training and inferencing, video transcoding and other similar applications. Barr states that the G4 instances will be available in multiple sizes and in bare-metal form. In addition, Amazon EKS introduced Kubernetes API Server endpoint access control. Kubernetes API Server endpoint access control will allow traffic between Kubernetes worker nodes, the Kubectl command line tool and the EKS-managed Kubernetes API server to stay within your Amazon Virtual Private Cloud (VPC). Google Cloud Google revealed its Stadia cloud gaming service at the Game Developers Conference (GDC) in San Francisco March 19. Stadia will stream games from the cloud to the Chrome browser, Chromecast and Pixel devices. Google power this streaming using its own data centers, which it says will “ensure servers are as close to players around the world as possible” and deliver lower latency to stream games effectively across the internet. Google Cloud is also the official cloud provider of the NCAA March Madness basketball tournament. Google Cloud is teaming up with student developers to analyze data and implement prediction analysis of games in the tournament using cloud tools. Microsoft Azure Last week, Azure announced in coordination with NVIDIA’s GTC event ONNX runtime integration with NVIDIA. This announcement is a step towards open and interoperable AI, letting developers easily leverage industry-leading GPU acceleration regardless of their choice of framework. Developers can now use TensorRT through ONNX Runtime to accelerate inferencing of ONNX models, which can be exported or converted from PyTorch, TensorFlow and many other popular frameworks. Other Cloud News Huawei announced a new event on March 29 at the ISP Asia Pacific Summit 2019 Conference in Bali, Indonesia. The event’s theme is slated to be “Leading new ICT – Accelerate Service Innovation with ISP.” |