On June 1, at the Alibaba Cloud Summit Guangdong-Hong Kong-Macao Greater Bay Area, Ding Yu, general manager of Alibaba Cloud's intelligent cloud-native application platform, announced that the serverless container service ASK has been fully upgraded to further help enterprises and developers reduce costs and improve efficiency.
Gartner has predicted that by 2023, 70% of AI applications will be developed based on container and serverless technologies. As an important technical component of cloud native, K8s has been widely recognized by developers and enterprises, but its own complexity and steep learning curve are still daunting.
Alibaba Cloud released the first serverless container service ASK in 2018. Its essence is to decouple the runtime of the container from the specific node operating environment, allowing users to directly deploy applications without managing K8s nodes and servers, greatly reducing the cost of container services. usage threshold. Currently, ASK is widely used in scenarios such as containerized applications, online business elasticity, and AI/big data computing tasks.
In this release, ASK further integrates the capabilities of Alibaba Cloud infrastructure, bringing significant improvements in usage costs, creation efficiency, compatibility with heterogeneous resources, and flexible supply guarantees. It meets the new demands arising from the explosion of AI scenarios.
Ding Yu introduced that this ASK upgrade covers multiple dimensions, including full hosting support for components, the ability to pull AI images in seconds, and reduces the cost of use for customers. Specifically:
Fully managed components, zero K8s operation and maintenance costs: ASK has newly added hosting support for more than ten K8s core components such as CoreDNS and Metrics Server, providing dynamic capacity planning capabilities, customers do not need to deploy and maintain themselves, and minimize the complexity of operation and maintenance. At the same time, ASK has also added intelligent risk identification capabilities to support automatic upgrades of K8s versions to avoid application failures or abnormal risks caused by upgrades.
Extreme elasticity, accuracy rate increased by 80%: ASK provides the world's first container image cache-based AI large image pull capability in seconds, and AI application startup time is reduced by 90%. It also provides end-to-end elastic acceleration, full-stack optimization for AI/big data workload containerization, and improves access performance by 30% through data set acceleration;
In addition, this ASK also enhanced the ability of intelligent elastic prediction AHPA, compared with manual configuration, the elastic accuracy rate increased by 80%; at the same time, it added support for GPU.
Inclusive computing power, 40% price reduction: In order to provide customers with better services, release technical dividends, and make computing power more inclusive, ASK has added support for U instance specifications, and supports multiple processors in a unified manner. Instance price reductions of up to 40%.
Added the SavingPlan elastic version, which is oriented to the application of non-fixed peak and trough scenarios, and the additional cost is optimized by more than 10%. In order to further make prices more transparent, support for cost suites has been added to provide a clear insight into the cost of elastic resources and make cost management more convenient.
Take Shuhe Technology as an example. This is a company that provides efficient smart retail financial solutions for financial institutions. It has high requirements for the computing power of the model, including computing speed, accuracy of computing results, and real-time computing data.
The current problem is that the underlying application resources supporting model calculations cannot adjust machine resources to support computing capabilities according to the amount of requests. This is also a pain point that needs to be solved urgently in the process of rapid business development. At the same time, with the increase in the number of model online reasoning services, Shuhe's model services have become increasingly large, bloated, and difficult to manage. This situation not only leads to waste of resources, but also increases maintenance and upgrade costs.
In order to solve these "stubborn problems", Shuhe Technology adopts Alibaba Cloud ASK to deploy online models without K8s node management, dynamically uses POD according to real-time traffic, and saves resource costs by 60%; through the ASK Knative service, it solves the grayscale of Shuhe models Coexistence of publishing and multiple versions; thanks to the advantages of ASK's automatic scaling and shrinking to 0, operating costs are reduced and service availability is greatly improved.
"Using Alibaba Cloud container service Knative and ECI virtual nodes to cooperate and deploy, while ensuring the stability of the online model to cope with sudden traffic, it also significantly improves resource utilization efficiency and greatly saves resource costs." Shuhe Technology AI Experiment Zhou Weipeng, person in charge of the AI platform of the laboratory, said.
In order to allow container developers and users who are interested in using Kubernetes to deploy AI model services to better experience ASK, Alibaba Cloud has launched a new "Easy Deployment of Enterprise-level Stable Diffusion Based on ASK" scenario experience. Deploy the Stable Diffusion service that meets the enterprise-level elasticity requirements through Knative, and experience the elasticity of ASK through the pressure test experiment of the service.
Experience URL: https://developer.aliyun.com/adc/scenario/de33e7d3065949f3b81db292b2dca5ea
In order to let more developers feel the charm of serverless technology, Cloud Native Application Platform and Tianchi jointly launched the 2023 Cloud Native Programming Challenge. Over the past eight years, more than 50,000 teams have participated. There are many excellent contestants and outstanding works emerging every year, and the Cloud Native Programming Challenge has become the technology vane in the cloud native field.
This year's competition is divided into three tracks, which solve the problems often encountered in different scenarios, including serverless cold start, plug-in design in the field of application security, and designing an innovative application through SAE. The competition is about to start, with a cash prize of 360,000 yuan, so stay tuned!
ASK free trial gameplay and then upgrade
Currently, ASK has joined the Alibaba Cloud Feitian Free Trial Program to provide developers and enterprises with a certain amount of free trial resources. Create a Kubernetes cluster in 3 minutes and start the journey of container elasticity.
Game 1: ASK developer evaluation is officially launched
In order for you to experience the capabilities of ASK products more quickly and conveniently, after receiving the trial resource package, you can choose any one of the following two given scenarios to fully experience the advantages of ASK products in specific applications, and experience ASK around Process evaluation:
- Evaluation address:
https://developer.aliyun.com/mission/review/ask
First prize: 1 best review, get Redmi Watch 3 + developer review a full set of customized peripherals (mouse pad, frisbee, canvas bag, Yunxiaobao) + Alibaba Cloud community high-quality evaluation certificate + Alibaba Cloud community home page expert display for a week ;
Second prize: 5 high-quality reviews, and get Alibaba Cloud customized backpack + developer evaluation limited release Yunxiaobao doll + Alibaba Cloud community high-quality evaluation certificate.
Play method 2: Scene experience: Realize barrage service in ASK
In order to let everyone experience more ASK capabilities, a special experience scene is set up - "Implementing barrage services in ASK". In this experience, the system automatically generates an ASK cluster to provide a business operating environment. Send barrage messages to HomePage through the front end, and then HomePage sends the barrage information to message processing for processing. After the processing is completed, the page will display the barrage results obtained by the front end.
- Experience address:
https://help.aliyun.com/document_detail/612667.html
More gameplay and surprise gifts are available at the ASK product upgrade conference, click here to enter the live broadcast room.