Skip to main content

Week 14

Tuesday​

No meeting held

Thursday​

🎥 Meeting Recording​

🚩 Agenda​

  1. Presentation by Brennan Macaig on the role of an SRE
  2. Break for Pizza 🍕
  3. Break into teams and work on Immersion

🪄 Meeting Resources​

TBD

📓 Meeting Notes​

Attendance:

  • In-Person: 15
  • Virtual: 2

Presenter: Brennan Macaig - Site Reliability Engineer

Site Reliability Engineering - an Introduction

  • Why should you listen?
  • What is an SRE?
    • Differences to DevOps?
    • Why care about SRE?
    • How is SRE implemented?
  • What is it like?
  • At Scale
  • AWS
  • Stories
  • Questions

Why Listen?

  • Experience with GCP & Azure
  • 5 Years of experience in the field
  • Worked at small and large companies
  • Went to UML (2017-2020)
    • Was a college tour guide and RA for a while
  • Used to a full stack dev

What even is SRE?

  • SRE are software engineers
  • Solve operational problems
  • Cross between traditional IT and Software development

Why Care?

  • Limit the impacts of problems on customers
  • Customers obsessed people
  • Help maintain the SLA (Service level agreement)
  • Improve observability into the tech stack
    • Catch errors before impacting revenue
  • Co-own productions with developers
  • Highly focused on observability
    • Telemetry
    • Documentation
    • Visibility of information

Why should I care?

  • Rewarding jobs that re relatively high paying
  • Fast paced
  • Working on unique and unsolved problem
  • Real impact into customer experiences

What makes an SRE an SRE?

  • Be an owner
  • Value customers above all
  • Implement the DevOps philosophy, without being DevOps
  • Create tools

Thinking like an SRE - DevOps

  • 5 Principles
    • Reduce org. silos
    • Accept failure as NORMAL and INEVITABLE
    • Implement chain gradually
    • Leverage tooling and automation
    • Measure everything
  • SRE is not DevOps

Thinking like an SRE - Not DevOps

  • Own production with devs.
  • Make gradual changes and canarying and testing
  • Leverage/create tooling and automation, especially to eliminate toil, tech debt, and overhead.
  • Toil - "Toil is the kind of work tied to running a production service that tends to be manual, repetitive, automatable, tactical, devoid of enduring value, and that scales linearly as a service grows"

Terms

  • SLIs
    • Service level indicator
    • Measurement of services health or status.
  • SLOs
    • Service Level Objectives
    • A target for your service regarding availability.
  • SLAs
    • Service Level Agreements
    • A promise made to customers about availability.

Day in a life

  • On-call and production outage management
  • Writing code
  • Standup, sprints, scrum, agile, JIRA tickets, investigations
  • Meetings
    • Meet to discuss new services
    • Meet to discuss old services
  • WRiting terraform
  • Operations tasks
  • Deal with scaling services
  • Write techn plans
    • Reduce techn debt
  • Write scripts to reduce daily toil
  • Other infra provisioning tools
  • Deal with CI/CD pipelines

Customer Obsession

  • Focus on positive customer experiences
  • Developer's customers are the people who use the company's app
  • SRE customers are developers
  • Make processes better for devs
  • Make workflows dev centered
  • Make doing things wrong more challenging
  • Protect customers from themselves
  • Support when customers do custom stuff

At Scale

  • Services at scale need to do?
    • Reliability
    • Easy to fix and work on
    • Easy to deploy and create
    • Be process defined/constrained
    • Be monitored
    • When possible, self heal
      • Avoid late night pages

AWS - What do SREs touch?

  • Since SREs own production, touch all prod assets
  • Company dependent
  • Typical things Brennan uses
    • EC2, EKS, IAM, Lambda, S3, SNS, RDS, KMS, Route53, APIGateway, Step Functions, Cloudwatch, WAF, SecurityGroups (IAM)

❓ Questions​

5 Years into the industry and no problem?

  • Yes
  • Backstory: @ UML for 2017 - 2020, not great in school, COVID happened, didn't enjoy online school, got internship. Stayed on Fall semester, over winter break decided to stay full time. Obtained a full time offer, continue working. Got job at Draft Kings after original company went down. Started to Draft Kings, then moved to Aqua Security, now a new company.
  • Most important thing is getting internship experience. Internships and Co-op are very important to get hired.

Still important to get a degree

  • Easier to get a degree is you have a degree or still working on a degree

Work Life balance

  • Work to live
  • Hard during on-call weeks
  • Being able to seperate your work life and
  • Company culture has more to do with the amount of worklaod you get as an SRE than the role itself.
  • Depends more on your ability to segement or avoid work in your personal time and vice versa

What is ownership:

  • Tagging and ensuring compliance
  • Responsibility for when things go wrong, how to keep improving

🧑‍💻 Hands on Section​

No hands on section today, good luck on exams!

🚀 Next Meeting​

We'll be having a entertaining and fun night of AWS Trivia with prizes for winners! Be sure to check back on the schedule for more details soon!

🥂 Cheers!