HighLoad++ Moscow 2018
| title | author | length |
|---|---|---|
| HighLoad++ Moscow 2018 Opening (November 8) | 22m | |
Opening session of the HighLoad++ Moscow 2018 conference, a major professional event for developers of high-load systems. It provides basic information about… | ||
| HighLoad++ 2018 — Day One and the Awards | 02m | |
A report on the HighLoad++ conference series for developers of high-load systems, including details about the Moscow 2018 event (Nov 8–9) and the HighLoad++… | ||
| HighLoad++ 2018 Live Stream. November 8, Congress Hall | Vadim Madison, Ivan Kruglov, Nikolay Golov, Dmitriy Stolyarov, Nikolay Sivko, Aleksandr Krasheninnikov, Aleksey Lesovskiy, Anton Povarov | 07h 30m |
A one-day program listing HighLoad++ Moscow 2018 talks about building and operating high-load systems: microservices (what we know), challenges of… | ||
| HighLoad++ 2018 Live Stream. November 9, Congress Hall | Vladimir Yants, Oleg Bartunov, Fedor Sigaev, Ilya Kosmodemyanskiy, Yuriy Nasretdinov, Aleksey Kirpichnikov, Aleksey Ivanov | 06h 53m |
The schedule lists talks from HighLoad++ Moscow 2018 covering scalable systems and databases: testing a monolith for hundreds of client versions, engineering… | ||
| Be Your Own HR: What and How to Develop | Takhir Bazarov | 58m |
Explains how recruiting and people-development roles differ in software companies and questions whether a dedicated HR is always necessary. Covers which… | ||
| What Do We Know About Microservices? | Vadim Madison | 36m |
Discusses which metrics are collected from microservices during build, testing, staging and production, naming concrete data points used to detect misbehavior… | ||
| The Thorns of Containerized Applications and Microservices | Ivan Kruglov | 53m |
Booking.com describes its evolution through three generations of private cloud platforms: an early Mesos/Marathon deployment (used about six months), a… | ||
| One Approach to Implementing Data Discovery in a Microservice Architecture | Nikolay Golov | 51m |
Describes a Data Discovery solution for microservice architectures using Database-per-service and polyglot persistence. Covers challenges of distributed data… | ||
| Databases and Kubernetes | Dmitriy Stolyarov | 55m |
Flant shares practical experience running databases and stateful applications on Kubernetes, using concrete examples to show when it makes sense to deploy… | ||
| Apache Kafka as a Foundation for Building Bicycles | Nikolay Sivko | 54m |
Describes building a specialized time-series database on top of Apache Kafka to meet high-load requirements for performance, low latency, and storage… | ||
| Speeding Event Processing to 1.6M/sec: Badoo’s Experience | Aleksandr Krasheninnikov | 47m |
Description of migrating a near-realtime event-processing pipeline at Badoo from a Spark+Hadoop stack to ClickHouse to handle much larger volumes. Covers the… | ||
| Top Development Mistakes When Working with PostgreSQL | Aleksey Lesovskiy | 53m |
A PostgreSQL consulting team from Data Egret reviews the most common mistakes developers make when working with PostgreSQL, illustrating how development… | ||
| "Platform" at Badoo: How We Built Infrastructure Development | Anton Povarov | 41m |
Badoo describes building an internal infrastructure platform to handle recurring large tasks like deployment, photo/video storage and delivery, metrics… | ||
| A Monolith for Hundreds of Client Versions: How We Write and Maintain Tests | Vladimir Yants | 52m |
Badoo engineers describe their testing practice for a large monolithic backend that must support hundreds of client versions. They explain their development… | ||
| How to Create High Load Out of Thin Air | Oleg Bartunov, Fedor Sigaev | 56m |
The talk examines what people mean by “highload” in web and enterprise systems, arguing that many so-called highload incidents indicate underlying design or… | ||
| How to Become a Great Database Specialist | Ilya Kosmodemyanskiy | 42m |
Overview of database technologies used in high-load projects, covering traditional systems (MySQL, PostgreSQL), legacy/enterprise ones (DB2) and various NoSQL… | ||
| Query Performance Analysis in ClickHouse | Aleksey Milovidov | 57m |
Covers diagnosing slow ClickHouse queries by showing where to look in execution plans and resource usage. Describes techniques and tools to analyze whether a… | ||
| How VK Inserts Data into ClickHouse from Tens of Thousands of Servers | Yuriy Nasretdinov | 41m |
VK's use of ClickHouse for storing logs and statistics at large scale, including how much data is kept and how it is written from tens of thousands of servers… | ||
| A Custom Incident Notification System | Aleksey Kirpichnikov | 46m |
A team built an in-house alerting system for emergency situations that decides who to wake up at night. The presentation explains handling noisy or missing… | ||
| Dropbox’s Traffic Infrastructure | Aleksey Ivanov | 48m |
Overview of Dropbox’s traffic infrastructure from user request to application servers, covering external DNS/BGP load balancing using RUM and global… | ||
| ClickHouse Is Slowing Down | Kirill Shvakov | 53m |
A practical guide showing how to build inefficient solutions with the ClickHouse columnar database. It demonstrates specific anti-patterns and mistakes when… | ||
| Yandex.Metrica and a Non‑standard ClickHouse | Aleksandr Makarov | 55m |
Yandex.Metrica’s analytics team describes running a heavily customized ClickHouse deployment they adopted in 2014, explaining why ClickHouse was chosen and how… | ||
| "Delayed Data" — Our Mechanism for Ensuring Consistency | Andrey Litunenko | 47m |
Describes a migration from a homemade data bus to Apache Kafka and Go for inter-service data exchange to solve message loss and single-threaded import… | ||
| What to Do If You Need to Process a Billion Chaotic Tasks per Day in PHP | Anton Gorin | 49m |
Describes how ManyChat processes hundreds of millions of users and over 400,000 businesses on Facebook Messenger by triggering immediate business-logic… | ||
| Yandex.Cloud: The Hard Choice Between Time-to-Market and Product Manageability | Yan Leshchinskiy | 33m |
Yandex.Cloud's engineering choices for building a public cloud platform: which existing Yandex components were reused (for example, their internal object… | ||
| Rewrite or Not? Life with Technical Debt | Anton Potapov | 53m |
A retrospective on managing technical debt in a production system, describing practical changes made: removing VPN to improve developer ergonomics, introducing… | ||
| Golang: Specific Performance Issues | Daniil Podolskiy, Kirill Danshin | 34m |
Shows Go performance pitfalls and optimization techniques for high‑load systems: discusses Go runtime behavior (garbage collector, goroutines, scheduler)… | ||
| Building Our Own Browser from Scratch. Part One: HTML | Aleksandr Borisov | 52m |
Describes building a custom, high-performance HTML parser and DOM implementation from scratch, including how a “real” HTML parser differs from simpler parsers… | ||
| DNS at Facebook | Oleg Obleukhov | 44m |
Explains how Facebook uses DNS as part of its load-balancing and global service infrastructure: how DNS resource records are generated and propagated into… | ||
| Solving High-load Load Balancing Problems | Aleksey Burylov | 35m |
Describes high-load load-balancing problems encountered at Qiwi and various techniques to increase fault tolerance. Presents qiwi-thrift-pool, a load-balancing… | ||
| Replicated service mesh: hardening systems against failure modes in load balancing | Oleg Klyudt | 51m |
Practical recipe for preventing mesh meltdown: prefer L7 load balancing so the proxy can do content routing, traffic splitting and per-request failover; derive… | ||
| Solving Service Auto-scaling in Yandex.Cloud | Ryurik Krylov | 47m |
Yandex.Cloud's team describes building an automation service for deploying and horizontally autoscaling virtual machines. The talk covers the technologies used… | ||
| Databases in the Cloud | Vladimir Borodin | 40m |
Overview of Yandex’s cloud database infrastructure: describes the internal compute platform that runs most stateless services, existing MapReduce and object… | ||
| Yandex’s Crowdsourcing Ecosystem | Olga Megorskaya | 38m |
Yandex describes its large crowdsourcing ecosystem that scales human work across tasks like data collection for machine learning, content generation, manual… | ||
| How Clear Task Estimation Materializes Results | Aleksandr Pavlyut | 54m |
The talk examines how clear, object-oriented problem estimation drives successful realization of ideas by focusing on the concept of object complexity — the… | ||
| Feature Toggles, or How to Deploy Features Without a Release | Yan Ashenkampf | 51m |
Describes using feature toggles (feature flags) to deploy a new personalized product-recommendation feature for an online store without waiting for a formal… | ||
| What to Do When a Minute of Downtime Costs $100,000 | Evgeniy Kuzovlev | 51m |
Describes techniques for making financial-transaction systems survive extreme availability requirements when a minute of downtime costs very large sums. Covers… | ||
| Best Practices for Native Cloud Services | Elena Grakhovats | 52m |
Explains what requirements a microservice must meet to be considered cloud-native and where those requirements come from, using the example of turning a simple… | ||
| Metrics! Metrics! Metrics! | David O'Brien | 48m |
Covers using Microsoft Azure’s metrics capabilities to monitor application health and understand environment behavior. Explains which Azure metrics and tools… | ||
| PostgreSQL 11 and Beyond: Overview of New Features and Trends | Ivan Panchenko | 50m |
Overview of PostgreSQL 11 features and the longer-term projects and trends behind them; covers specific new capabilities introduced in PostgreSQL 11 and… | ||
| Scaling PostgreSQL Replicas Under Load | Andrey Borodin, Vladimir Leskov | 38m |
Techniques and practical tweaks to speed up PostgreSQL backups, presented with focus on cloud-deployed clusters. Covers WAL-G-based backup mechanisms… | ||
| Find Me a Job: How Search Works on hh.ru | Aleksey Bichuk | 50m |
Evolution of hh.ru job search from a simple Lucene-based implementation to a high‑load search cluster with personalization using machine learning, quality… | ||
| Multi‑criteria Optimization of Search Results at Avito | Andrey Drozdov | 50m |
Avito handles hundreds of millions of search queries daily and faces the problem of satisfying multiple user groups whose ranking preferences conflict. The… | ||
| USE, RED and Golden Signals for Performance Analysis and Optimization | Petr Zaytsev | 51m |
Explains three performance-analysis and troubleshooting methodologies—USE (Utilization, Saturation, Errors), RED (Rates, Errors, Duration), and the Golden… | ||
| MariaDB and MySQL — What Statistics the Optimizer Uses | Sergey Golubchik | 42m |
Explains which table and column statistics MariaDB and MySQL optimizers use to estimate query costs and choose execution plans. Covers types of collected… | ||
| FAQ on the Architecture and Operation of VKontakte | Aleksey Akulovich | 49m |
Overview of VKontakte’s backend architecture and operational practices: how servers interact, where and why they still use “plain” PHP and which other… | ||
| Fighting for Survivability Under DDoS | Georgiy Tarasov, Artem Gavrichenkov | 54m |
Describes designing DDoS-resilient services by giving systems characteristics that let them keep operating as attackers change attack method, vector, and… | ||
| Technical Aspects of Internet Blocking in Russia | Filipp Kulin | 47m |
Explains the technical details of internet blocking in Russia: how the blocking mechanism is organized (who, what, where, when, how), why Roskomnadzor performs… | ||
| How to Take a Backup in a Distributed System Without Anyone Noticing | Ivan Rakov | 56m |
Describes how GridGain (built on Apache Ignite) handles creating backups of large, constantly modified distributed datasets—up to 200 TB across 160… | ||
| Choosing a Replication System for PostgreSQL | Viktor Egorov | 48m |
Compares PostgreSQL replication systems and how to choose among them for different needs like high availability, backup, and scaling when data won’t fit on one… | ||
| Unconventional Uses of ClickHouse | Aleksandr Zaytsev | 44m |
Covers unconventional uses of ClickHouse (the open-source columnar DBMS from Yandex), showing how its flexibility lets engineers apply it beyond typical log… | ||
| Scale Your Metrics with Elasticsearch | Philipp Krenn | 52m |
Shows concrete scaling choices: a 3‑node Docker Compose cluster with 3 primary shards, zero replicas for the demo, per‑minute indices for the demo ingest, and… | ||
| Ranking Listings with Machine Learning and ElasticSearch | Pavel Tarasov | 39m |
Describes how cian.ru implemented personalized ad ranking for search by combining ElasticSearch with complex machine‑learning models (gradient boosting and… | ||
| How to Defeat Brute Force and Leave Hackers with Nothing | Aleksey Ermishkin | 43m |
Discusses why users keep choosing weak passwords (like “love”, “god”, “sex”) and how database breaches expose most user passwords even when salted hashing is… | ||
| Recent Changes in the Linux I/O Stack from a DBA’s Perspective | Ilya Kosmodemyanskiy | 48m |
Covers recent redesigns of the Linux I/O stack that affect database administrators, explaining which kernel I/O technologies (e.g., block layer, io_uring… | ||
| eBPF for Linux Performance Analysis | Petr Zaytsev | 48m |
eBPF as a modern interface for instrumenting the Linux kernel and applications: what eBPF is, how it works, and the new capabilities it enables for performance… | ||
| Load Testing Atlassian Jira | Aleksey Matveev | 51m |
A technical session on load testing Atlassian Jira, describing how Jira was subjected to performance and stress tests. It covers the testing tools and… | ||
| Camunda on Microservices | Aleksandr Trekhlebov | 54m |
How to move from a monolithic BPM system to a microservices architecture using Camunda. Covers distributing processes across microservices, coordinating… | ||
| A 4K Streaming Platform for a Million Concurrent Viewers | Aleksandr Tobol | 48m |
Describes Odnoklassniki’s video streaming platform that delivers over 3 Tbit/s for 4K and Full HD live streams to millions of viewers. Covers the engineering… | ||
| How we deliver 1Tbps of video with our WebRTC peer-to-peer network | Jordan Pittier | 46m |
Streamroot’s WebRTC DNA delivers 1.2 Tbps peak of peer-to-peer video and offloads about 70% of CDN traffic by matching viewers via a router/tracker system… | ||
| Almost No Magic, or How to Smoothly Deliver a Terabit Video Stream | Mikhail Raychenko | 44m |
Describes delivering terabit-scale live video streams with a practical, low-magic engineering approach: how to distribute extremely large-bandwidth video… | ||
| A High‑load Distributed Control System for a Modern Nuclear Power Plant | Vadim Podolnyy | 01h 00m |
A new distributed control platform for modern nuclear power plants is described, including hard real-time coordination of over 150 specialized subsystems that… | ||
| Jedi Techniques in Team Management | Viktoriya Yurkevich | 50m |
Connects common engineering-manager habits (preferring technical tools over people) to ineffective team management and argues that standard coaching… | ||
| Why CTOs Should Code | Aleksandr Ziza | 52m |
Explains why a CTO should continue coding: argues that hands-on programming helps lead high-load system projects, improves managerial decision-making, and… | ||
| # Tools for Managing a Large Team | Dmitriy Bezuglyy | 51m |
Discusses tools and practices for managing engineering teams larger than ~30 people, addressing leadership challenges beyond the typical 7±2 team size. Covers… | ||
| Describing Infrastructure in Terraform for the Future | Anton Babenko | 50m |
Covers how to structure infrastructure-as-code using Terraform, discussing recurring problems teams face because there aren’t established best practices… | ||
| From Ansible to SaltStack — How and Why | Dmitriy Samokhvalov | 51m |
The talk compares using Ansible vs SaltStack for deploying and managing infrastructure at Alfa-Bank: it describes problems encountered with Ansible, what… | ||
| BeyondCorp: A DevOps Security Model Without Sign‑On and VPN | Gleb Lesnikov | 48m |
Describes implementing BeyondCorp-style zero-trust access for DevOps to remove VPNs and centralized credential management. Covers problems of many servers and… | ||
| A Survival Guide to MongoDB | Sergey Zagurskiy | 48m |
Practical guide to operating MongoDB under production load: covers common pitfalls that appear as traffic grows, which aren’t obvious at start, and describes… | ||
| The cost of MongoDB ACID transactions in theory and practice | Henrik Ingo | 49m |
Крупнейшая профессиональная конференция для разработчиков высоконагруженных систем Saint HighLoad++ 2026
Подробнее: https://clck.ru/3QZHTb
Июнь… | ||
| Building a 100‑Petabyte Analytics Warehouse | Aleksandr Mazurov | 49m |
Criteo describes building one of Europe’s largest Hadoop clusters with Hive as the central data-processing tool, explaining how their Hive platform evolved… | ||
| Continuous Optimization for distributed BigData analysis | Kai Sasaki | 48m |
Platform ingests about 1 million records per second and processes trillions per day using Hadoop and Presto. It replaces Parquet with a messagepack-based… | ||
| Attracting tinkoff.ru | Aleksandr Polomodov | 51m |
Presentation describes Tinkoff.ru’s system for acquiring new customers, defining what they consider “effective acquisition” and how that concept led to a… | ||
| Data Engineers and Who Needs Them | Valentin Gogichashvili | 48m |
The talk discusses why large organizations need data engineers and what problems they solve when companies collect and store large volumes of data. It covers… | ||
| How We Optimize Processes in Custom Software Development | Kirill Vetchinkin | 48m |
A team that builds custom financial systems for telecom and banking describes how they optimize business and technical processes to run project-after-project… | ||
| Evolution of the Moscow Exchange Trading and Clearing System Architecture | S.Kostanbaev | 50m |
Describes the evolution of the trading-and-clearing system architecture at the Moscow Exchange, tracing its development from a simple monolithic system in the… | ||
| How We Implemented the Core of Alfa‑Bank’s Investment Business on Tarantool | V. Drynkin | 49m |
Describes how Alfa-Bank rebuilt the core of its investment business using Tarantool as the database/engine. Covers the legacy architecture from the 2000s, the… | ||
| Medical Chatbot and a Physician Rating System | Ilya Larchenko | 50m |
Describes building two data-science products trained on patient medical records: a chatbot that collects patient complaints and performs initial routing, and a… | ||
| ‘Vertical Reality Glasses’: Fixing Typos in Search Queries | Tigran Saluev | 49m |
Describes building a typo-correction system for search queries, focusing on handling frequent user misspellings and service-specific vocabularies. Covers why… | ||
| Billing at DrimSim | Dmitriy Simonov | 46m |
Describes the MVNx billing system built for virtual operator Drimsim, detailing a microservices-based architecture, the specialized databases used, and how… | ||
| Deploying MySQL on Kubernetes & Openshift | Aleksandr Rubin | 56m |
Shows a working deployment that brings up a three‑node Percona XtraDB (Galera) cluster plus ProxySQL and monitoring with one deployment flow (Helm or oc… | ||
| "Charge Up" or CDC from MariaDB and Postgres into MariaDB Columnstore Analytics | Roman Nozdrin | 37m |
Demonstrates Change Data Capture (CDC) pipelines to stream data from MariaDB Server 10.3 and PostgreSQL 10 into the MariaDB ColumnStore analytical engine… | ||
| MySQL 8.0: SQL and NoSQL Scalability | Vittorio Cioe | 50m |
MySQL 8.0 adds a document store on top of InnoDB so JSON is stored in binary form with automatic validation and path access, collections map to two-column… | ||
| Lessons from Migrating a 10 TB PostgreSQL Cluster to AWS | Aleksandr Kukushkin | 49m |
A presentation about migrating a large PostgreSQL cluster (10 TB) used at Zalando to AWS, covering challenges of supporting over 700 clusters with mixed… | ||
| Experiments with Postgres in Docker and the Cloud | Nikolay Samokhvalov | 56m |
Examines running PostgreSQL in Docker containers and in cloud managed services, comparing technologies like AWS RDS, Google Cloud SQL, and Kubernetes operators… | ||
| Tools for Creating PostgreSQL Backups | Andrey Salnikov | 49m |
A practical overview of PostgreSQL backup tools and strategies, comparing logical backups (pg_dump) and binary backups (pg_basebackup) plus built-in and… | ||
| Serverless – are you ready for the revolution? | Michał Jankowski | 51m |
Explains how Azure serverless actually behaves: use Azure Functions with event queues to move from a single prototype function to a resilient pipeline of seven… | ||
| Replication Across Different Database Engines | Mikhail Buylov | 23m |
Describes moving parts of MySQL functionality into other systems and where MySQL’s capabilities fall short. Compares database caching in memcache with using… | ||
| Tarantool 2.1 Adds SQL Support: Details | Kirill Yukhin | 49m |
Tarantool 2.1 adds SQL support compliant with ANSI-style syntax. It describes how Tarantool executes nontrivial SELECT queries and implements data-integrity… | ||
| Make Your Database Dream of Electric Sheep: Designing for Autonomous Operation | Andy Pavlo | 53m |
Research prototypes show concrete gains: auto-tune using Gaussian processes on GPflow/TensorFlow automatically recommends knob sets for Postgres/MySQL… | ||
| Will Postgres Live Forever? | Bruce Momjian | 52m |
Argues that Postgres survives because open source follows a different lifecycle than proprietary software, with 32 years of development and a cadence of annual… | ||
| BBM’s 150M+ users Oracle to Postgres migration without downtime | Álvaro Hernández | 42m |
BBM migrates a multi-terabyte Oracle auth/metadata cluster to Postgres on GCP for a 150M+ user service, using ora2pg for DDL, Oracle GoldenGate for CDC initial… | ||
| Achieving Sub‑Millisecond Response Times in a Java Trading System on Linux | Aleksey Ragozin | 52m |
Describes how to build a Java-based trading system on Linux that achieves sub-millisecond response times. Covers specific techniques and optimizations for… | ||
| How We Crawl 60 Million Web Pages a Day | Aleksandr Sibiryakov | 50m |
A team built a web-crawling content system for a search engine that visits 14 million domains and fetches up to 100 pages per domain per month, then re-crawls… | ||
| How We Launched Russia’s First Kubernetes‑as‑a‑Service in Mail.Ru Cloud | Dmitriy Lazarenko | 48m |
Mail.Ru Cloud Solutions describes launching Russia’s first managed Kubernetes-as-a-Service offering: they explain using Kubernetes as the orchestration… | ||
| Function as a Service in private cloud | Sergey Rybalkin | 51m |
Alibaba private cloud implements an internal FaaS platform built in Kotlin to cut developer overhead, keep functions small (ideally ~80 lines) and provide a… | ||
| Choosing the Right Tool to Deploy Containerized Applications in the Cloud | V.Pantyukhin | 38m |
Compares AWS container orchestration services Amazon ECS and Amazon EKS (Kubernetes) and explains when to choose each for deploying and scaling containerized… | ||
| Optimizing Kubernetes Resource Requests/Limits for Cost-Efficiency and Latency | Henning Jacobs | 47m |
Collected metrics from 80+ clusters reveal a large fraction of pods run without proper memory or CPU requests, causing scheduling slack, unexpected OOM kills… | ||
| Automating Disk Replacement with Ansible | Artem Aleksandrov | 46m |
Automating hard-drive replacements using Ansible: the talk explains how to automate the routine process of swapping failed disks in large server fleets (tens… | ||
| Building the Next-Generation Backend on FoundationDB | Oleg Illarionov | 50m |
Presentation explains building a next-generation backend using FoundationDB, a sorted key–value database with ACID transactions. It discusses how FoundationDB… | ||
| A School for Programmers as a Hiring Solution | Lev Ekasov | 50m |
A company describes using an internal programming school to ease developer hiring shortages by training candidates. The talk explains the hiring problem—demand… | ||
| Zabbix: 100k NVPS on a Single Server | Mikhail Makurov, Maksim Chernetsov | 34m |
Zabbix monitoring platform is examined with focus on its limits and how to scale the analytics storage and optimize the Zabbix server and UI. The talk explains… | ||
| Monitoring for Developers! Technology for the Community! Profit for Everyone! | Vladimir Kolobaev | 52m |
Describes how Avito’s engineering team monitors a large distributed platform with over 5 million incoming metrics per minute and about one million business… | ||
| Big-Data Processing of Social Media User Data in AlibabaCloud MaxCompute | Maksim Alekseev | 51m |
Demonstrates combining heterogeneous user data for 700 million accounts from 15 sources, including AliExpress customer internal data and external… | ||
| Hadoop at Scale: We Built a Large Cluster, How Do We Maintain It? | Sergey Korsakov | 50m |
Describes operating a large Hadoop cluster after deployment: maintaining reliability, meeting client needs, and preserving the ecosystem at scale. Covers… | ||
| HighLoad++ 2018 Awards Ceremony | 30m | |
A ceremony presenting the HighLoad++ 2018 Awards recognizing contributions to Russia’s web development ecosystem. It lists event details (June 2026, Saint… | ||
| Rapid Development of Software Solutions with C++ and Lua | Aleksandr Borgardt | 47m |
Shows how to combine C++ and Lua to develop production services quickly while keeping scalability, fault tolerance, and low API latency. Describes using Lua… | ||
| Effil: An Alternative Approach to Multithreading in Lua | Mikhail Kupriyanov, Ilya Udalov | 38m |
Effil presents a different approach to implementing multithreading for the Lua language, comparing it with existing Lua concurrency solutions. The talk… | ||
| The ML Lifecycle in Combat Conditions | Sergey Vinogradov | 52m |
Covers practical lifecycle of machine learning in production, arguing that model training is only about a quarter of the work while most effort goes to data… | ||
| MyRocks deep dive and production deployment at Facebook | Yoshinori Matsunobu | 51m |
MyRocks (a RocksDB-based MySQL storage engine) and its production deployment at Facebook are examined, covering features added such as Blind Deletes and TTL… | ||
| Demystifying MySQL Replication Crash Safety | Jean-François Gagné | 48m |
Covers concrete failure modes and exact configs that break or restore replication crash safety in MySQL 5.5–8.0: 5.5 is unsafe; 5.6 adds… | ||
| High Loads in the Industrial Internet of Things | Evgeniy Potapov | 35m |
An overview of Industrial Internet of Things (IIoT) technologies and the challenges of operating them under high load. Covers industrial IoT device classes… | ||
| Building a High-Load Cloud Call Center | Evgeniy Pinashin | 50m |
Explains how to scale an Asterisk-based telephony service when a single Asterisk server cannot handle required concurrent calls (e.g., 300 vs 1000+). Describes… | ||
| Teaching a Machine to Synthesize Paracetamol | Artem Kondyukov | 51m |
Describes using machine learning to 'teach' a system to synthesize paracetamol (acetaminophen). Covers applying ML techniques to a problem hard to formalize… | ||
| Forecasting E‑Commerce Sales with Gradient Boosting | Aleksandr Alekseytsev | 54m |
An automatic warehouse restocking system at OZON.ru is driven by a gradient boosting ML model for sales forecasting. The talk covers problem setup and… | ||
| Tips & Tricks for Fast Neural Net Inference in Production | Dmitriy Korobchenko | 51m |
Check speed early during model construction and training and prefer simple architectural moves that actually cut work: use 1x1 convolutions to cheaply reduce… | ||
| How Neural Networks Helped Graphics | Evgeniy Tumanov | 50m |
The talk explains how neural networks were applied to graphics and visual effects production, using examples like large-scale rendering workflows (e.g., Weta… | ||
| Women’s Networks: How Neural Networks Help the Beauty Industry | Artem Prosvetov | 42m |
Neural networks applied in the beauty industry: how ML models are used to predict human behavior for commercial gain. Covers using data-science approaches and… | ||
| Prod, Do Stack: Production-Ready Solutions via Data Science Competitions | Andrey Kiselev | 33m |
The talk describes Dbrain’s new infrastructure for running machine-learning competitions where participants submit executable source code. Models are trained… | ||
| Phishing: The War of Infinity | Evgeniy Bogomazov | 55m |
Covers phishing and social engineering as ongoing threats to stealing user logins and passwords, including attacks on banks, crypto exchanges, and political… | ||
| Modern fraud trends in RTB | Ekaterina Safonova | 45m |
Company processes ~6 billion daily requests and finds mobile apps now near 50 percent of traffic so app fraud surges. Detection uses a three stage waterfall… | ||
| The Role of Row Level Security in High-Load Projects | Aleksandr Tokarev | 43m |
Examines where and how to implement row-level security in high‑load projects, discussing whether row-level access control is necessary and comparing available… | ||
| How Tinkoff.ru Uses Greenplum | Dmitriy Nemchin | 52m |
Describes how Tinkoff.ru runs Greenplum for large-scale analytics: they implemented a custom replication system from Oracle and Postgres, run about 4,000 daily… | ||
| How We Built Our Own Netfilter with Intel DPDK and Prefix Trees | Aleksandr Samoylov | 55m |
Describes building a custom packet-filtering engine based on Intel DPDK and prefix trees as an alternative to Linux Netfilter. Covers using DPDK for… | ||
| Decentralized Technology for Bypassing Firewalls | Evgeniy Safronov | 41m |
Describes peer-to-peer connectivity problems caused by NATs and provider firewalls for decentralized P2P apps, and presents techniques to traverse these… | ||
| QUIC illustrated | Aleksandr Krizhanovskiy | 54m |
Explains what QUIC actually does and what was tested: observed Internet share ~7–10% and Google ~35% QUIC traffic, captured a Chrome/alt‑svc session and… | ||
| Software Defined Storage the Linux way | Philipp Reisner | 48m |
Linux combines LVM (with thin provisioning), dm‑raid, dm‑cache, targets/initiators and DRBD into a coherent stack and Linstore adds a single control plane that… | ||
| Serious Retail | Aleksandr Lishchuk, Dmitriy Tsvetkov | 44m |
A case study of evolving a classic 3‑tier OLTP/ERP system at X5 Retail Group during rapid organic growth, describing performance optimization techniques for a… | ||
| AWS Cost Reduction - Experiences and Strategies | Andrew Boag | 49m |
Catalyst documents concrete practices: they cut AWS spend by 30–40% reduction through audits, tagging and account separation, pragmatic use of reserved… | ||
| Replication in Tarantool: Configuration and Usage | Georgiy Kirichenko | 51m |
Covers replication in the Tarantool database: how to configure replication and how to use it to provide high availability by replicating servers or forming… | ||
| VShard — horizontal scaling in Tarantool | Vladislav Shpilevoy | 47m |
VShard describes a horizontal scaling solution for the Tarantool database, contrasting it with the earlier Shard module. It explains VShard’s approach to… | ||
| Proper database staging for Postgres | Nikolay Samokhvalov | 57m |
Covers proper staging environments for PostgreSQL databases: how to set up and manage staging that reflects production, handle schema and data synchronization… | ||
| A Kubernetes cluster on your laptop: introduction to minikube | Viktor Safronov | 01h 44m |
An introduction to running a local Kubernetes cluster using minikube: what minikube is, how to install and configure it on a laptop, and how to create and… | ||
| Load testing a message bus | Sergey Zhurin | 43m |
Describes load-testing a message bus where input load is not constant or stepwise but changes nonlinearly over time. Covers why standard load-testing tools… | ||
| Load-fire exercises on queries | Anton Barabanov, Dmitriy Godov | 48m |
Overview of techniques used for load-testing and those observed in real web attacks, focusing on methods for generating and handling high-load request floods… | ||
| ClickHouse: blitz meetup (questions and answers) | Aleksey Milovidov | 53m |
A brief Q&A session about ClickHouse covering practical questions and answers for developers using the ClickHouse columnar OLAP database. The meetup focuses on… | ||
| Sharing Lua game-logic code between server and client | Andrey Trifanov | 52m |
Describes using the same Lua game logic on both server and client by running OpenResty (Lua) on the backend and the Defold 2D engine (which also supports Lua)… | ||
| How we work on the stability of our Lua implementation | Anton Soldatov | 48m |
IPONWEB describes maintaining a custom fork of LuaJIT used for over ten years to run business logic, focusing on ensuring the runtime’s stability. The talk… | ||
| Multiple inheritance in Lua | Sergey Lelyakov | 37m |
Explains using component-based multiple inheritance in Lua to improve code reuse in large, evolving RTB (real-time bidding) platforms. Describes applying Lua’s… | ||
| DNS at Facebook | Oleg Obleukhov | 44m |
Overview of how Facebook uses DNS for traffic balancing and load distribution, including the role of Facebook’s DNS infrastructure in that process. Explains… | ||
| Video report on | Andrey gevara | 04m |
A video report about the HighLoad++ professional conference for high-load system developers, covering the Moscow 2018 edition (November 8–9) and mentioning the… | ||
| nothing matches | ||