On October 19th, the 2021 Yunqi Conference with the theme of “Frontier · Exploration · Imagination” opened in Hangzhou. The conference lasted for four days and was back offline and open to the public for free. According to media reports, at today’s conference,Alibaba Cloud released a new operating system “Dragon Lizard” and announced its open source.At the same time, the operating system laboratory of Ali Dharma Academy was also announced.
access:
2021 Tmall Double 11 Red Packet Collection Page
According to reports,The dragon lizard operating system is positioned on the server side and supports multiple chip architectures and computing scenarios such as X86 and ARM.
The system has been polished within Alibaba for 10 years and has effectively supported Tmall Double 11 over the years. Its performance and stability have withstood severe tests.
In addition, Longli has made multiple optimizations for cloud-native application development,It brings a 40% overall performance improvement for typical scenarios on the cloud, a 50% reduction in failure rate, compatibility with the CentOS ecosystem, supports one-click migration, and provides full-stack national secret capabilities.
It is reported that the dragon lizard operating system is completely open source, providing services through open source communities and operating system vendors.
In the future, Alibaba Cloud plans to invest 2 billion yuan in special funds for Dragon Lizard, and unite 100 ecological partners to promote ecological construction and provide technical support for at least ten years.
It is worth mentioning that, in addition to the dragon lizard, Alibaba Cloud also launched the fourth-generation Shenlong architecture.
This is also the new generation of Feitian Cloud operating system virtualization technology, and for the first time it is equipped with the world’s only large-scale elastic RDMA acceleration network.The computing architecture innovation brought by Dragon 4.0 brings cloud computing into the era of 5 subtle delays.
According to reports, in the deep learning scenario, the fourth-generation Shenlong can improve the training performance of distributed NLP and visual computing by 30%. In the big data scenario, it can improve Spark’s computing performance by 30%.
In the database scenario, MySQL performance can be increased by up to 60%, and Redis hybrid read and write throughput can be increased by 130%. The throughput performance per second of NginxSSL connection is increased by 420%.
.