Highload fwdays'19
2019-10-05
Kyiv, Ukraine
who it's for
backend engineers, devops engineers, site reliability engineers, data engineers, machine learning engineers, database administrators, engineering managers, tech leads
topics
serverless applications with GraphQL; AWS Lambda for PHP; GitLab and Docker build optimization; monolith to microservices migration; continuous delivery for machine learning; Kafka stream processing; Apache Beam and Cloud Dataflow; Neo4j graph databases; SQL firewall design; autoscaling and cost optimisation; Google Anthos with Istio; designing for catastrophic failure
| title | author | length |
|---|---|---|
| Scaling ML from 0 to millions of users | Julien Simon | 44m |
Compares three concrete deployment paths: single VM with deep learning AMI for quick notebooks, container clusters on ECS/EKS with CPU and GPU nodegroups… | ||
| Practical Experience Using AWS Lambda in High-Load PHP Applications | Yevgen Lysenko | 49m |
A PHP ticketing platform handling ~1.5–2M visits/month moved parts to AWS Lambda using Bref to run PHP 7.2/7.3. The talk covers practical constraints (cold… | ||
| Practical Tips for Build Optimization for GitLab and Docker Users | Dmytro Patkovskyi | 36m |
Practical build-optimization rules for teams using GitLab CI and Docker: treat artifacts, cache, and persistent volumes as different mechanisms (artifacts only… | ||
| Solving Scalability by Converting a Monolith into Microservices | Artem Nikulchenko | 43m |
Describes converting a monolithic app into targeted microservices running on Kubernetes in Google Cloud to improve scalability. Specific changes include… | ||
| 100 releases per day | Dmytro Voloshyn | 41m |
Highload fwdays’20 conference will be held on October 03, 2020 in Kyiv, Ukraine 🇺🇦
Register here: https://fwdays.com/event/highload-fwdays-2020
⬇️⬇️⬇️ Useful… | ||
| Database Highload: When to Scale and When to Optimize | Denis Reznik | 49m |
Shows how to decide between scaling and optimizing using concrete metrics (CPU%, requests/sec, disk latency/IOPS, cache life) rather than buzzwords. Describes… | ||
| Continuous Delivery for Machine Learning — The Future of MLOps | Kyrylo Perevozchykov | 39m |
Describes concrete MLOps patterns and a practical CI/CD approach for ML: three deployment modes, a model registry with blue‑green swaps, shadow and… | ||
| Developing Serverless Applications with GraphQL | Marcia Villalba | 36m |
Claims that serverless GraphQL removes boilerplate get tested: a live example wires AWS AppSync to a DynamoDB table and a Lambda, deploys via Serverless… | ||
| Panel Discussion: Architecture | Dmytro Lavrinenko, Maksym Bezuglyi | 42m |
The panel argues for hiring software architects early and having them fluent in code and business trade-offs; one company described an architect training… | ||
| Panel discussion "Tech / Team Leads" | Dmytro Nemesh, Illia Seleznov, Kirill Latysh, Dmytro Voloshyn | 48m |
Highload fwdays’20 conference will be held on October 03, 2020 in Kyiv, Ukraine 🇺🇦
Register here: https://fwdays.com/event/highload-fwdays-2020
⬇️⬇️⬇️ Useful… | ||
| Panel discussion "DevOps" | Oleksandr Milushev, Mikalai Alimenkou | 46m |
Highload fwdays’20 conference will be held on October 03, 2020 in Kyiv, Ukraine 🇺🇦
Register here:
https://fwdays.com/event/highload-fwdays-2020
⬇️⬇️⬇️… | ||
| Panel Discussion: Databases | Denis Reznik, Alexander Solovyov, Andrey Varenyk, Nedashkivskyi Yevhen | 34m |
Panelists compare concrete tradeoffs for databases: MongoDB transactions add latency and split‑brain risk, so teams sometimes replace MongoDB when they need… | ||
| ETL is Dead. Long Live Streams with Apache Kafka | Taras Klioba | 40m |
Highload fwdays’20 conference will be held on October 03, 2020 in Kyiv, Ukraine 🇺🇦
Register here: https://fwdays.com/event/highload-fwdays-2020
⬇️⬇️⬇️ Useful… | ||
| Neo4j for Solving Graph-Based Problems | Andrey Varenyk | 38m |
The talk explains how Neo4j was used to replace a two-hour exhaustive route enumeration with a production pipeline that delivers a median ~340 ms search time… | ||
| Autoscale — Pay When You Earn | Kirill Latysh | 45m |
Designs architecture around unit economics and a single operational metric correlated with delayed revenue, then autoscale to that metric instead of guessing… | ||
| Default to Open: Creating a DevOps Culture | Katrina Novakovic | 32m |
Argues that DevOps succeeds only when culture changes, not just tools, and gives concrete starting points: measure the four Accelerate metrics including… | ||
| From encoding to machine learning a story of managing media data at Netflix | Arsen Kostenko | 29m |
Netflix builds a media intelligence platform that turns ingestion into a managed, versioned dataset so models actually work in production: initial duct‑taped… | ||
| Complex Application Design | Orkhan Gasimov | 41m |
Presents a practical methodology illustrated by three side projects and a production stack using Node.js (vanilla) with Docker and Jenkins, frontend in… | ||
| Kafka Deployment at Scale | Oleksandr Nitavskyi | 35m |
Criteo describes operating Kafka at massive scale: 10 million messages/sec (~500B/day) and ~180 TB/day compressed across ~14 clusters with 72‑hour retention… | ||
| Building an SQL Firewall: Insights from Developers | Artem Storozhuk | 35m |
A SQL firewall implemented as a proxy parses incoming SQL into ASTs, matches queries against policy trees to block or rewrite them, and provides logging, TLS… | ||
| Multi-Cloud with Google Anthos and Istio — How to Stop Worrying About Infrastructure | Anton Grishko | 47m |
Google Anthos is used to centralize configuration for many Kubernetes clusters and large developer fleets by keeping config in a single Git-backed location and… | ||
| Advances in Stream Analytics: Apache Beam and Google Cloud Dataflow deep-dive | Sergei Sokolenko | 47m |
Clear claims and numbers: Apache Beam + Google Cloud Dataflow use a backend shuffle and separate state storage from compute so large-grouping jobs drop from… | ||
| Release After a Year of Development | Dmitriy Menshikov | 44m |
A B2B real‑time video platform released a rushed December build and spent ~40 days chasing intermittent stream drops. Engineers used strace, perf, flamegraphs… | ||
| The role of catastrophic failure in software design | Jeff Atwood, Michael Krakovskiy | 44m |
Argument: failures teach more than wins, supported by actionable practices and tools. Claims: run post-mortems with timelines, keep unit tests run on every… | ||
| Highlights | Jeff Atwood | 03m |
Conference highlights present a sequence of concrete takeaways not generalities: a short report that StackOverflow and Discourse people find the event well… | ||
| nothing matches | ||