Systems @Scale Winter 2022
as the organisers put it
Systems @Scale Winter 2022 is a technical conference designed for engineers that manage large-scale information systems serving millions of people. The operation of large-scale systems often introduces complex, unprecedented engineering challenges. The @Scale community focuses on bringing people together to discuss these challenges and collaborate on the development of new solutions. The Winter 2022 series […]
who it's for
platform engineers, site reliability engineers, infrastructure engineers, cloud engineers, devops engineers, incident responders, security engineers
topics
utilization saturation metrics; memory vs CPU tradeoffs; push-based async schedulers; Tailor adapters for streams; serverless workflow building blocks; torrent-based package distribution; chaos-controller Kubernetes operator; kubernetes chaos experiments; internal cloud single-pane control; demand efficiency portal and quotas; RAMP-TAO atomic visibility; Terraform as a service (TFaaS)
| title | author | length |
|---|---|---|
| Effective Resource Management: Through Systems Design | Salim Virji | 18m |
Example system is an image server that serves cat photos and hits <10% CPU while maxing RAM so the team compresses in memory to trade CPU cycles for higher… | ||
| Asynchronous Computing at Scale: Overview and Learning | Sayak Kundu, Artem Denisov | 22m |
Meta reports the platform handles more than a trillion asynchronous workloads per day after re-architecting from a pull-based dispatcher reading Fox (Facebook… | ||
| Network Workflow Automation at Scale | Nathan Sullivan, Nikita Vetoshkin | 18m |
Meta describes a workflow platform that runs thousands of concurrent workflows and ~100,000 executions per month by composing reusable binary building blocks… | ||
| Chaos Engineering at Scale | Nikos Katirtzis | 19m |
Expedia Group builds a chaos platform for a common runtime on Kubernetes serving 15,000 applications across tens of thousands of nodes, integrating the open… | ||
| Building Meta's Internal Cloud: the opportunities and challenges | Surupa Biswas | 19m |
Meta runs an internal cloud that serves 3.7 billion people on millions of machines and handles Tau’s quadrillion queries per day, using systems like twine for… | ||
| Live Q&A Session – Anca Agape | Anca Agape | 27m |
Q&A lists concrete practices and hard lessons: use the utilization saturation errors metrics to get cost‑per‑request; prefer queues for command buffering and… | ||
| Managing Demand and Efficiency for Meta Multi-tenant Services | Tina Luo, Sherry Chen | 20m |
Meta builds a one-stop Demand Efficiency Portal that enforces quotas and admission control, runs automatic regression detection with FB Detect, and surfaces… | ||
| RAMP-TAO: Layering Atomic Transactions on Meta's Online TAO Data Store | Audrey Cheng | 21m |
Describes RAMP-TAO, a layered protocol that adds atomic visibility to Meta’s TAO by using the existing refill library for limited multiversion metadata and… | ||
| Security Culture & Scaling Your Security Humans | Paul Hawkins | 20m |
AWS embeds security into builders by training over 2,000 software engineers as Guardians who act as local appsec proxies, uses an automated ticketing-first… | ||
| Managing Incidents | Ramón Medrano Llamas | 27m |
Google-style incident process that assigns Incident Commander, Operations Lead and Communications/Planning roles, uses runbooks and BigQuery log queries, and… | ||
| Enabling Infrastructure as Code | Rahul Raja, Wensi Hu | 22m |
LinkedIn runs an internal Terraform as a service (TFaaS): remote execution with Azure blob storage, service-bus queues and workers running Terraform binaries… | ||
| Live Q&A Session – Liane Praza | Liane Praza | 29m |
Live Q&A surfaces concrete practices and results: ownership is annotated by inserting wrapper function C between callers so cost attribution moves without… | ||
| nothing matches | ||