Jiekun’s Resume #

About Me #

I am a backend engineer with a passion for open-source software. Throughout my career, I have gained extensive experience in developing applications for e-commerce advertising and logistics. As time has progressed, my focus has naturally shifted towards infrastructure, specifically in the realm of observability.

My expertise as a backend engineer encompasses a diverse range of skills, including proficiency in the Go programming language, database, caching, and microservice governance. I continually strive to stay at the forefront of industry trends and emerging technologies, currently honing my knowledge in observability domain such as OpenTelemetry, eBPF, and Prometheus.

Outside of work, I actively engage in technical writing and knowledge sharing. Additionally, I have had the opportunity to present at CNCF events like KubeCon + CloudNativeCon and KCD, where I contributed my insights and experiences to the community.

Work Experience #

• Building a PromQL Proxy to safeguard infrastructure like Thanos and Prometheus, while also auditing query operations.
• Utilizing eBPF to collect data and transform it into correlated metrics, logs, and traces.

Currently, our eBPF collector ingests millions of spans per second to provide correlated R.E.D (Rate, Errors, and Duration) metrics, logs, and traces to tens of thousands of applications, utilizing only 20 CPU cores and 60 GiB of memory on the server side.

• Evaluating and migrating our Prometheus & Thanos setup to VictoriaMetrics. I designed a query traffic comparison solution based on the PromQL Proxy I built. With this, we can verify whether VictoriaMetrics provides similar query results and faster response times.

We have successfully migrated our test cluster to utilize VictoriaMetrics, which supports the same data collection volume of 700k samples per scrape as before, but at 50% of the resource cost. We expect to complete the migration within 6 months, which will save us $4,000+ per month.

• Maintaining the OpenTelemetry Collector for Traces processing. I shared my experience with Span Metrics Connector at KubeCon + CloudNativeCon Observability Day.

• Leading the data migration from TiDB to MySQL, including technical design, sharding planning, dual-write implementation, and data validation.
• Responsible for distributed tracing integration, providing simplified integration through various middleware extension plugins for the team. I accumulated extensive experience in distributed tracing SDKs and sampling and shared them at KubeCon + CloudNativeCon + Open Source Summit China.

Off-Hour #

Learning #

The knowledge I have acquired in the field of computer science has been continually changing over time, while my enthusiasm for this domain has remained unwavering.

There are several examples of how I have dedicated my spare time to learning, striving to improve myself and those around me:

• I learned how to build complex and large-scale tests, particularly end-to-end (e2e) tests, when I submitted my first Pull Request for Kubernetes.
  • This experience inspired me to introduce a simplified version of e2e testing into my work and lead others in its adoption.
• I learned how to design highly modularized systems when I submitted my first Pull Request for OpenTelemetry Collector.
  • This experience inspired me to redesign and implement the Exporter Interface for the DeepFlow project, and to shared the concept of modularized design with others.

Currently, I remain actively engaged in the OpenTelemetry and other observability communities to submit fixes and features to the upstream.

Writing #

Beyond the open-source community, technical writing has been my primary means of sharing knowledge with others. My blog serves as a platform to showcase the expertise I have accumulated in various domains over the past few years, including:

• Persistent Data Structures in VictoriaMetrics (Part 1): vmagent. (3 May. 2024). https://jiekun.dev/posts/vmagent-data-structures/

• KubeCon Europe 2024: Cloud Native In La Ville-Lumière | In The Spotlight. (23 Mar. 2024). https://jiekun.dev/posts/kubecon-eu-2024-ii/

• KubeCon 2023 Shanghai: A Recap. (05 Oct. 2023). https://jiekun.dev/posts/kubecon-2023/

• Real-Time Checking System. (17 Mar. 2022). https://jiekun.dev/posts/data-check/

• Paper Reading: HotRing. (13 Mar. 2021). https://jiekun.dev/posts/hotring/

• Paper Reading: Let’s Talk About MVCC. (20 Nov. 2020). https://jiekun.dev/posts/mvcc/

  More

Event Engagements #

Public Talks #

TTChat’s Story: Connect Metrics, Logs and Traces with eBPF

eBPF 的上层应用：让 Logging、Metrics 和 Tracing 联动起来

Balancing Cost and Quality in OpenTelemetry: An Evaluation of Sampling Policies

Internal Talks #

I have given an average of 5 technical talks each year within the organizations I have worked for, since 2020.