Eric Wang · Infrastructure engineer
I build infrastructure where latency, security and uptime all matter.
Infrastructure engineer in Sydney. About twenty years (since 2006) in production, from enterprise Linux and data-centre operations to cloud-native DevOps and SRE. Writing about reliability, Linux performance and the trade-offs between them; learning notes are kept apart from established practice.
- ~20years in production
- AWSprimary cloud platform
- Linux → Cloud → SREcareer path
- 30+tools in stack
Latest writing
All writingFrom the data centre to cloud SRE: what changed in twenty years, and what didn't
Twenty years from racked servers to cloud infrastructure: the parts of the job that changed completely, and the fundamentals that never moved.
Stack
Technology I have worked with, grouped by area.
- Linux and performance
- CPU affinity, NUMA, HugePages, IRQ tuning, Kernel & network tuning, Latency measurement
- AWS and infrastructure as code
- AWS, EC2, VPC, IAM, EKS, S3, CloudWatch, Route 53, RDS, Terraform, Ansible
- Containers and delivery
- Kubernetes, Helm, Argo CD / GitOps, Jenkins, Bamboo, GitHub Actions
- Observability
- Prometheus, Alertmanager, Grafana, Loki, ELK
- Data and messaging
- PostgreSQL, MySQL, Redis, Kafka, RabbitMQ, Spark, ClickHouse
- Security
- IAM, Access controls, OS hardening, Network connectivity
- Scripting
- Python, Bash
Lab
Learning notes, kept separate from established production experience.
From the LabAll Lab notes
LAB-001 Lab
Learning notes: GPU/HPC infrastructure, and how it differs from ordinary cloud instances
An ops engineer's first map of GPU clusters: capacity, drivers, interconnects, topology, storage, schedulers, monitoring, failure modes and cost. Learning notes, not experience.
Contact
A question about a post, a correction, or a collaboration idea? Get in touch.
Open to consulting and part-time work.