李麒傑(James Li)
Worked on: Linux, Container, Kubernetes, Ceph, OpenStack
Programming: Golang, Shell(Bash)
(+886)978-132-500 [email protected] Ci-Jie cijie-li
Experience at Work
ASUS
Open Cloud Infrastructure Software (OCIS) Team - Software Engineer • Sep 2018 - Now
Work Experience：
- Experienced in building a mass storage system with total capacity of more than 45 PB, which provides the storage service for OpenStack (around 450 nodes) and Kubernetes (around 250 nodes). IOPS: Write(30GB/s) and Read(17GB/s).
- Integrated a Simple Queue Service & Simple Notification Service service, named Triton, which is compatible with AWS standards. It can deal with requests up to 7,000 per second.
- Designed and developed a Cloud Storage Service that is compatible with AWS S3 standard. It not only can use lots of backend storage such as NFS or HSM (ASUS Business Cloud Storage) also support third-party tools like S3 Browser, Cyberduck and SDK.
- Integrated Triton SQS with OpenStack Barbican to provide mechanism of Server Side Encryption.
- Integrated Triton SNS & SQS with ElasticSearch for metadata aggregation.
- Integrated lots of softwares such as DHCP, Apache, Kickstart (for CentOS) and Pressed (for Ubuntu) to design a PXE solution that can install over 100 computers automatically at one time.
- Integrated our application with K3s to build a SIT testing environment. It provides a Kubernetes based on container for our application to create resources on it like PV, PVC for testing...
- Integrated Athens with our private gitlab to provide a Go module Proxy for all of our R&D engineers.
- Designed an automated building solution for development environment based on container, which is simply used under offline.
- Designed a tool for quickly building binary files of Go language based on container.
- Designed Go module which can call Linux kernel function (statx) to acquire ctime(create time) of file.
- Developed LDAP Interface for ASUS Identify Service. It can integrate with IoT devices which authenticate by LDAP protocol.
Survey open source and practice in own team：
- Integrated Ceph RGW with OpenStack Keystone - The user information will be managed by OpenStack Keystone.
- Integrated Ceph RGW with OpenStack Barbican - The object in server side can be encrypted by using secret which is stored in OpenStack Barbican.
- Integrated Ceph RGW with ElasticSearch - The object metadata can be stored in ElasticSearch and look for any metadata quickly by using ElasticSearch query scripts.
- Implemented Ceph Multiple Side mechanism to prevent data broken in single side.
Experience about AWS Services
- Surveyed and organized AWS S3 API documents and choose 23 APIs of them to practice in ASUS CloudStorage Service.
- Implemented ASUS CloudStorage Service which is compatible with AWS S3 standard.
- Surveyed and organized AWS SNS/SQS API documents and choose 17 APIs of SNS and 8 APIs of SQS to practice in ASUS Triton Service.
- Refer to AWS SNS/SQS documents to implement ASUS Triton Service and provide email, HTTP/HTTPs ...etc for notification.
- Implemented a S3 benchmark tool for testing performance about multipart upload operation.
- Published paper for 2017 2nd International Conference (ICMSSP 2017): Designing A Simple Storage Services (S3) Compatible System Based on Ceph Software-Defined Storage System
Contribution for open source on Github
I contribute lots of open source in my free time. Focused on correcting mistakes in documents which I used at work. Including:
- k3s-io / helm-controller
- Kong / docs.konghq.com
- Azure / multi-cluster-networking
- kubesphere / openelb
And the others are implemented by myself:
- Ci-Jie/goproxy: GoProxy is a module proxy server for golang. It provides a datastore to store your private go packages from gitlab.
- Ci-Jie/s3bench: S3bench is a tool that is responsible for performance testing, including lots of S3 operations such as PutObject, GetObject and MultipleUpload.
Topic Sharing in Public Community
SDN & Cloud Native Meetup #18
Topic: How to implement a S3 API gateway based on Ceph RGW and analysis performance • Jul 2019 - Speaker
Taichung City Free and OpenSource Software Fans Community - TFC
Topic: Blockchain & Docker & Kubernetes • Mar 2018 - Speaker
SDN & Cloud Native Meetup #4
Topic: Build the Blockchain as service (BaaS) Using Ethereum on Kubernetes • Apr 2018 - Speaker
Education
National Taichung University of Science and Technology - Emerging Smart IoT Cloud Computing System Lab
Master • Sep 2016 - Aug 2018
Master's Thesis: Deep Learning Based Auto-Scaling Load Balancing Mechanism for Distributed Software-Defined Storage System
Oral Presentation (2017, August). Designing A Simple Storage Services (S3) Compatible System Based on Ceph Software-Defined Storage System. In Proceedings of the 2017 2nd International Conference on Multimedia Systems and Signal Processing (pp. 6-10).
Development Items：
- Developed and designed a smart monitor system and portal in own lab.
- Implement a vote system based on blockchain software - Ethereum.
Survey open source project and implement in own lab：
- Implemented a HA Kubernetes cluster in our smart lab and let other teams can launch their services on it for testing.
- Implemented Ethereum for our team members and taught them how to use and build it based on container and Kubernetes.
