dotScale 2014
who it's for
site reliability engineers, platform engineers, infrastructure engineers, DevOps engineers, backend developers, security engineers, storage engineers
topics
dynamic cloud application scaling; OpenZFS storage; application security and risks; extensibility as a product; PHP ecosystem and practices; Salt configuration management; Wikimedia-scale infrastructure; Iris cloud messaging; developer platform design; boot2docker Docker tooling; Consul service discovery; healthcare.gov operational lessons
| title | author | length |
|---|---|---|
| Everything I learned about dynamic scaling of cloud apps. | Laurence Moroney | 19m |
A live One Direction event drives massive ten‑minute spikes so the team uses App Engine for VM autoscaling but avoids constant capacity costs by writing… | ||
| OpenZFS | Matthew Ahrens | 16m |
ZFS claims end-to-end integrity by checksums all data and metadata and avoids repairs after crashes via copy-on-write atomic on-disk consistency, eliminating… | ||
| Security and Risks | Alison Gianotto | 22m |
Gives concrete metrics and pragmatic fixes: defines Mega breaches as >10M records and cites eight in 2013 versus one in 2012 — a 700% increase — and daily… | ||
| Extensibility as a Product | Jean-Michel Lemieux | 19m |
Argues that extensibility must be a first-class product feature and demonstrates a working prototype: browser-hosted add-ons stored in jgit repos, edited… | ||
| My Take on PHP | Fabien Potencier | 18m |
Claims that PHP performance can improve by an order of magnitude by avoiding per-request teardown and redoing bootstrap work, proposing to move bootstrapping… | ||
| Why did I create Salt? | Thomas Hatch | 19m |
Salt (SaltStack) claims concrete results: benchmarks dramatically faster than competitors, scales to tens of thousands of nodes, and uses an asynchronous… | ||
| The Wikimedia infrastructure | Faidon Liambotis | 16m |
Wikimedia runs its own mid‑size physical network of about a thousand servers across Ashburn and Dallas with caching POPs in Amsterdam and San Francisco and… | ||
| Iris: Simplifying cloud messaging | Péter Szilágyi | 06m |
Claims a drop in complexity by showing a runnable broker implemented in about 300 lines of Go, describes concrete wiring for HTTP and WebSocket clients, and… | ||
| Building the Ideal Developer Platform | Jeff Lindsay | 19m |
Claims a concrete platform: a meta-stack that runs on Docker and boots a container compute grid to bridge Heroku-style PaaS and host-based infrastructure. Core… | ||
| boot2docker | Steeve Morin | 05m |
Boot2docker is a minimal Linux distribution that boots Docker containers directly: a 25 megabyte iso that boots in three seconds and runs entirely from RAM. It… | ||
| Building Robust Systems With Consul | Mitchell Hashimoto | 17m |
Consul provides service discovery via DNS and HTTP API, uses a per-server agent that caches and serves stale data, and exposes a KV store with dynamic changes… | ||
| Today we make the Internet of tomorrow | Michiel Leenaars | 07m |
Organization hands out microgrants with the condition projects stay open source and funds standard work at the XMPP Foundation, the IETF and W3C. Funded… | ||
| Life in the Trenches of healthcare.gov | Robert Kennedy | 27m |
Outage stories focus on concrete failures not theory: a permissions change at 10:48 causes permission denied errors that instantaneously drop latency and make… | ||
| Darwin & Protocols | Paul Mockapetris | 28m |
DNS evolution faces concrete limits: cache poisoning exposed by Dan Kaminsky accelerates the need for DNSSEC is 15 years old and slow to deploy, many… | ||
| nothing matches | ||