Client cases > JD Cloud helps Xrender in building a new generation of cloud rendering platform
JD Cloud helps Xrender in building a new generation of cloud rendering platform

JD Cloud helps Xrender in building a new generation of cloud rendering platform

Industry involved:
Internet Application
Website link:
Click and enter
Product using:
Virtual Machines  Cloud Disk Service  Virtual Private Cloud  Load Balancer  Elastic Network Interface 
Application Type:
Rendering and High-performance Computing
Client cases

Customer Introduction

Jiangsu Cudatec Technology Co., Ltd. is a high-tech enterprise specializing in 3D visual computing cloud technology research, development and application. Currently, it independently runs a cloud rending platform, providing the users of traditional videos and animations as well as indoors architectural design with the self-help rendering service, the project management service, and the customized integrated solutions for 3D rendering. The platform (Xrender), since its formal outward operation in 2012, has been developing very rapidly, has a technical service team from the field of video and animation specialty, has rich experience in providing the rendering project service, and takes a leading position in the domestic cloud rendering service industry. The successful cases consist of the Monkey King: Hero is Back, Monster Hunt, Dragon Nest: Rise of the Black Dragon, Boonie Bears: Mystical Winter and other hot animation films, and more than 60% of film making companies producing films shown in the Chinese movie theaters have cooperated with Xrender.

Project Overview

Xrender is a leading domestic rendering service platform for graph cloud cluster and the self-service cloud rendering platform supporting on-line submission, self-service rendering, real-time monitoring and automatic download. This platform has a quite harsh requirement for hardware device, demanding powerful computing power and stability besides very high graphics performance. By deeply cooperating with Xrender, JD Cloud optimized in-batch creation of Virtual Machines, GPU Virtual Machines, large-capacity system disks, swift startup operating system and storage of cloud files, started up 300G of the system disk within a short period of time as per the customer’s business requirement, installed hundreds of Virtual Machines and GPU Virtual Machines of Windows System and business software, and at the same time read the material file of the file storage system. After completing rendering, the whole rendering platform automatically releases Virtual Machines, and stores the result in the Object Storage Service.

Solutions

After fully communicating with each other and carrying out tests, JD Cloud and Xrender have, according to the business features of Xrender and the characteristics of Xrender, worked out a solution complying with the featured rendering of cloud computing, which is detailed as follows:1. Virtual Machines expand elastically, and the rendering scenarios vary because of the customers served by Xrender. JD Cloud, according to the customer’s characteristics, specially designed this: Different computing resources of Virtual Machines in different numbers are called based on different business, the customer’s different time requirements and the historical data models, so that JD Cloud’s benefit in excellent performance is fully utilized to rapidly help the customer complete various rendering business. It can accomplish the rendering tasks from furniture effect picture to film and TV animations.2. As for selection of the optimal server, as the customer’s rendering scenarios are complicated, JD Cloud, according to different types of rendering, can call Virtual Machines and GPUs of different proportions by use of its benefit in the sufficient computing abilities of GPU and Virtual Machines, and by computation, choose the optimal proportion between performance and price to achieve the optimal selection of rendering cost performance, helping the customer complete the rendering task on the basis of saving every bit of money.3. Rapid large-scale startup of Virtual Machines: JD Cloud can, through the in-batch rapid startup function specially optimized for the characteristics of the rendering industry, simultaneously and successfully start hundreds of Virtual Machines within the specified time, rapidly improving the rendering business and saving every second of time for the customer.

Customer’s earnings

By cooperating with JD Cloud, Xrender has created a rendering platform based on JD Cloud, which brings significant benefits to the customer: 1 and save the investment of Xrender in IDC; traditional rendering platforms are based on physical servers and require the assistance from IDC, networking, electricity and other facilities as well as a large amount of startup capital; by cooperating with JD Cloud, Xrender saves a lot of capital and can invest more capital in research and development to create higher additional value. 2 and simplify the operation and maintenance: With the traditional Data Center not maintained and operated, lots of labor and materials are saved; the 24-hour uninterrupted operation of servers is guaranteed by professional JD Cloud, greatly reducing the pressure borne by Xrender’s IT basic operation and maintenance team; accordingly, Xrender can have the energy to focus on its own business making, enhancing its enterprise competitiveness. The flexible billing methods of 3 and help the customer decrease the financial pressure. As the resources of cloud computing are paid on demand, and under the principles of a higher fee for more use and a lower fee for less use, the input-output ratio is perfectly achieved based on different amounts of business.