Sydney Superuser Award Nominee: China Railway Corporation

Voting is now closed. Stay tuned to find out who wins!

It’s time for the community to determine the winner of the Superuser Award to be presented at the OpenStack Sydney Summit. Based on the community voting, the Superuser Editorial Advisory Board will review the nominees and determine the finalists and overall winner.

Now, it’s your turn.

The China Railway Corporation is among the seven nominees for the Superuser Awards. Review the nomination criteria below, check out the other nominees and rate the nominees before the deadline Wednesday, September 20 at 11:59 p.m. Pacific Time Zone.

Please specify the team and organization for nomination.

The team for nomination: China Railway Information Technology Center(CRITC)
The joint team: CRITC, Beijing SinoRail Information Technology Co. Ltd.(BJSRIT), Beijing T2Cloud Technology Co. Ltd., with a total of about 200 engineers.

CRITC:
Mingxing Gao/Project Director
Yang Liu/System Architect
Liang Liu/Senior Engineer

BJSRIT:
Wei Rao/R&D Manager
Guangqian Li/Technical Support Manager
Yahong Du/OpenStack Leader
Minhong Wang/IAD of OMS Group Leader
Gang Xu/MON & AMS of OMS Group Leader

T2Cloud:
Jinyang Zhao/R&D Manager
Yahui Hou/CMP Group Leader
Chao Xie/OpenStack Leader
Hanchen Lin/Testing Leader
Tony Xu/Software Architect

How has OpenStack transformed the organization’s business?

OpenStack is the key element to promote transforming the construction model of China railway information system from the traditional project-driven to platform-driven. It also supports the high-speed rail as the foundation of China’s growth strategy. The adoption of OpenStack marked the first time CRITC has fully and embraced open source technology with an open mind. The deployment of OpenStack cloud saved us millions of dollars and the application launch cycle has been shortened from several months to 1-2 days, with much quicker response to business requirements.Cloud computing has improved resource utilization rates, and measurements have shown that energy consumption has been saved by approximately 50 percent.

How has the organization participated in or contributed to the OpenStack community?

In 2014, CRITC started developing an open source cloud solution based on OpenStack and started to actively participate in OpenStack community activities from 2016 on. This included sharing a topic at the 2016 OpenStack Days China, two topics at 2017 Boston Summit, two topics at 2017 Sydney Summit to be held and giving a keynote presentation at 2017 OpenStack Days China. In 2016 and 2017 they also hosted OpenStack meetups in Beijing and Nanjing respectively. We have contributed 734 patch sets, 5,979 lines of code, and submitted and resolved 47 bugs for the OpenStack community. We have shared some practice experience of stress testing and performance optimization analysis at 2017 Boston summit and 2017 OpenStack Days China and written whitepapers for the communities.

What open source technologies does the organization use in its IT environment?

In addition to OpenStack, the China Railway Cloud depends on KVM, OpenVSwitch/LinuxBridge, Hadoop, Kafka, Flume, Spark, CentOS, LXC, Docker, Kubernetes, OpenShift, CEPH, GlusterFS, Redis, MongoDB, MySQL/MariaDB, Ansible, Open-Falcon, ELK, ZeroMQ/RabbitMQ, etc..

What is the scale of the OpenStack deployment?

Deployed about 5,000 physical server nodes, including about 800 KVM nodes and about 730 VMware nodes; 20PB SAN storage, 3PB distributed storage (Ceph). An additional 2,000 physical server nodes are to be deployed in the end of 2017..

Our OpenStack cloud platform, with a scale of 800 physical nodes, hosts thousands of VMs and a dozen of mission critical applications, which covers 18 railway bureaus and over 2,000 railway stations and powers the production. OpenStack cloud platform also well undertook the huge pressure brought by Spring Festival peak to the system, especially the over 31 billion daily average page view, and it has also supported stable, safe and 24/7 uninterrupted operation of real-time dispatching management for all the trains, locomotives and vehicles.

What kind of operational challenges have you overcome during your experience with OpenStack?

Suspension of OpenStack service: By monitoring the service process and the status of log generation, we can check whether the OpenStack service is hung up or not.

Version upgrade problems: Previously, we had already upgraded the cloud software from Essex to Juno version and from Juno to Liberty. With lots of changes made on the community edition, we are now upgrading the software from the Liberty version to Ocata and we will upgrade the production system online after the above work are done.

High availability of cloud: We realized there was a function of automatic failover to achieve high availability of VMs. Compute nodes would be isolated and their VMs would be evacuated to the other nodes in the same zone if a certain network is down or is unresponsive for two minutes.

How is this team innovating with OpenStack?

Comprehensive cloud solution plan: Based on open source components, we developed the Operation Management System (OMS) complementary to our cloud software. Currently OMS consists of monitor, automation and analysis.

Multidimensional and customer-oriented improvements: Modified front-end functions to optimize customer experience, added operation type logs for easier archiving by administrators, added permissions control, added failover function, etc..

Architecture adjustment for large-scale deployment: We verified that 800 servers hosting 100,000 VMs in the same region can be operating stably, supporting high availability of control nodes.

How many Certified OpenStack Administrators (COAs) are on your team?

There is currently one COA on our team. Our team plans to be more involved in the OpenStack community, and cultivate more talent, aiming to obtain 3-5 COAs in 2018 and 5-10 COAs in 2019.