Velocity 2016 Santa Clara
2016-06-20..2016-06-23
Santa Clara, United States
Velocity ·
[Velocity 2009, Velocity 2010, Velocity 2011, Velocity 2013, Velocity 2016 Santa Clara, Velocity 2016 New York, Velocity 2017]
who it's for
site reliability engineers, platform engineers, devops engineers, web performance engineers, security engineers, engineering managers
topics
Kafka-driven async pipeline; gatekeeper deploy systems; canary consumers on production traffic; dockerized workflow in production; initial congestion window TCP tuning; ad impact and tracking measurement; Snap telemetry with Grafana plugin; unreliable failure detectors for consensus; controlled consumption for microservices; DDoS mitigation and tabletop drills; .NET Core cross-platform tooling; disk-usage forecasting and heatmap UI
| title | author | length |
|---|---|---|
| Making bad ads sad. Rad! | Bruce Lawson | 02m |
Measurement shows ads make pages heavy: on top media sites 9% of page, 54% of load time and 55% of bandwidth while generating an average of 279 ad URL calls… | ||
| Building a system that never stops: New Relic at scale | Matthew Flaming | 11m |
New Relic handles over 200 services, more than 2 PB SSD and traffic growth that made horizontal sharding with a proxy tier fail by multiplying failure points… | ||
| Poised to deploy: The C-suite and adaptive capacity | Richard Cook | 02m |
Defines poised to deploy as knowing what the platform is supposed to do, how it actually works, what it’s doing now, and being able to change it, predict… | ||
| Jump-starting back-office connections | Harkeerat Bedi | 05m |
A CDN deploys a tool called Riptide that jump starts new persistent TCP connections by raising the initial congestion window based on a per‑destination moving… | ||
| Containers will not fix your broken culture, and other hard truths | Bridget Kromhout | 03m |
Claim: containers do not fix organizational culture or remove Conway’s law. Describes a small streaming site that ran Docker in production in October 2013 and… | ||
| My cloud makes pretty pictures | Matthew Brender, Raj Dutt | 06m |
Demo of Snap open telemetry framework collecting hardware counters through application metrics into a single API and metric catalog, paired with Grafana 3.0… | ||
| So we hear you like papers | Inés Sombra, Caitie McCaffrey | 04m |
Paper introduces unreliable failure detectors as an abstraction that decouples failure detection from implementation, defines two properties—completeness… | ||
| Tackling consumption chaos in microservice architectures | Mike Dvorkin | 06m |
Proposes a controlled consumption model that classifies consumers and maps them to either dedicated or shared clusters, isolating resource use and failure… | ||
| DDoS emotions | Artur Bergman | 13m |
A DDoS can hit like a targeted assault on infrastructure, customers and staff, not a routine outage: one example shows a sustained 160 Gbps flood with probing… | ||
| You got your design team in my security team! | Eleanor Saitta | 02m |
Argues that security lives at the system and human-impact level not the platform level, so systems not platforms is the guiding claim; notes lang sec doesn't… | ||
| “Microsoft is irrelevant to me. I use a Mac": Is this you? | Volker Will | 05m |
Microsoft describes a clear shift to open source and cross‑platform development, claiming over 7,000 Visual Studio extensions, .NET Core beyond Windows and… | ||
| Turning high-velocity data into leverage for people | Ozan Turgut | 05m |
Instrumented hosts get reduced to actionable signals: a simple forecast that divides remaining disk by last-hour usage flags three hosts likely to run out in a… | ||
| A young lady's illustrated primer to technical decision making | Charity Majors | 02m |
Claims that the rush to glue together third-party platforms and services massively raises system complexity, that many new solutions are not battle‑hardened… | ||
| Measuring what matters | Stephen Ludin | 05m |
Explains that the User Timing API lets developers record exact start and end points for critical sections so changes map to the things users actually notice… | ||
| Performance is about more than time series charts | Buddy Brewer | 06m |
Argues that fixing the slowest metrics wastes effort and recommends prioritizing pages by user impact: an x‑axis plot with green dots for response times and… | ||
| Building an open platform for government innovation | Diego Lapiduz | 04m |
cloud.gov builds a platform-as-a-service that turns source into runnable container images using buildpacks so developers only push code and worry about app… | ||
| Why DevOps is more than just automation | David Hayes | 05m |
Survey of a few hundred Ops and DevOps teams finds 85% missed critical alerts, DevOps teams respond faster (often minutes, typically under 30), yet the average… | ||
| The state of the W3C Web Performance Working Group | Todd Reifsteck, Philippe le Hegaret | 04m |
W3C Web Performance WG is defining concrete APIs and trade-offs: work targets memory measurement API that can report page, iframe or system memory while… | ||
| Assaf Resnick (BigPanda) | Assaf Resnick | 11m |
BigPanda presents a data‑science platform that ingests signals from 17 monitoring tools (Splunk, AppDynamics, New Relic, Datadog, SolarWinds, etc.), normalizes… | ||
| The relationship between DevOps and security | Elizabeth Lawler | 04m |
DevOps and security converge when security is treated as an ongoing engineering task rather than a break-fix mentality bolted on at deployment; bring… | ||
| Gabby Menahem (LoomSystems) | Gabby Menahem | 08m |
Claims an AI platform that parses logs and metrics with no configuration and returns human-like anomalies, trends and correlations; installs and yields a first… | ||
| Adopting a dockerized workflow | Michael Hausenblas | 05m |
Checklist for adopting a dockerized workflow: verify version control such as git and an automated CI/CD pipeline that actually produces images, pick which apps… | ||
| Bridget Kromhout (Pivotal) | Bridget Kromhout | 10m |
Containers do not magically fix organizational handoffs and incentives and require concrete practices to work in production. Demand that CI system is the only… | ||
| Shane Evans and Todd DeCapua (HPE) | Shane Evans, Todd DeCapua | 10m |
A short practical primer that claims to turn tribal knowledge into a repeatable practice, laid out as a 12-step path starting with culture, team and metrics… | ||
| Steve Kamerman (Scientia Mobile) | Steve Kamerman | 08m |
Device intelligence identifies a requesting handset from billions of requests per month using a compact device database (the WURFL data file) and drives an… | ||
| Mike Cohen and Mike Dvorkin (Cisco) | Mike Cohen, Mike Dvorkin | 09m |
Argues for policy-driven abstraction so services hide infrastructure, for removing linear team dependencies so changes stop rippling across orgs, and for… | ||
| Volker Will (Microsoft) | Volker Will | 06m |
Microsoft is shifting from Windows-only products to cross-platform development with Azure and Visual Studio Team Services at the center, offering extensible… | ||
| Jeff Geiser (Verizon Digital Media Services) | Jeff Geiser | 07m |
Verizon Digital Media Services combines the EdgeCast CDN and Uplink workflows into a platform that accepts single-file ingest, pushes content to an Anycast… | ||
| Tammy Everts (SOASTA) | Tammy Everts | 09m |
Collaboration with Google feeds 1.1 million sessions of real RUM data (stored in perpetuity by the vendor) into machine learning that examines 93 attributes… | ||
| Tom Lounibos (SOASTA) | Tom Lounibos | 09m |
Company describes a digital operation center that aggregates web and mobile user traces, correlates AppDynamics New Relic Splunk and Adobe Analytics feeds, and… | ||
| Lee Atchison (New Relic) | Lee Atchison | 08m |
Argues that scaling failures usually come from a tiny piece of logic whose complexity explodes and causes repeated outages (examples include recurring… | ||
| Miska Kaipiainen (Kontena) | Miska Kaipiainen | 07m |
Kontena delivers an open source, developer-focused container and microservices platform packaged as a CLI that provisions a runnable cluster on any cloud or… | ||
| Implementing security into an organization | Eleanor Saitta | 03m |
Most organizations treat security as bolted on or limited to testing and incident response, leaving almost no architectural security or formal security design… | ||
| Gareth Rusgrove (Puppet Labs) | Gareth Rusgrove | 07m |
Argument: adopting Google-born tech helps solve distributed problems but does not turn an organization into Google; many OSS projects copy Google patterns yet… | ||
| Ian Carrico (Vox Media) | Ian Carrico | 05m |
Vox Media builds a dedicated performance team in 2015 after years of rapid launches and a 2014 debt reckoning and focuses on reproducible testing tooling while… | ||
| nothing matches | ||