HighLoad++ 2017
2017-11-07..2017-11-08
Moscow, Russian Federation
[HighLoad++ 2012, HighLoad++ 2013, HighLoad++ 2014, HighLoad++ 2015, HighLoad++ 2016, HighLoad++ 2017, HighLoad++ 2019, HighLoad++ 2022, HighLoad++ 2023, HighLoad++ 2024, HighLoad++ 2025]
| title | author | length |
|---|---|---|
| Video Report from HighLoad++ 2017 | 02m | |
Video report and details about the HighLoad++ 2017 professional conference for developers of high-load systems, including locations and dates (Saint Petersburg… | ||
| Introduction to NGINX Unit | Igor Sysoev, Nikolay Shadrin | 36m |
Introduces the new open-source NGINX Unit application server (beta) — a next-generation architecture that serves as a flexible, efficient app server. Covers… | ||
| What's New in nginx? | Maksim Dunin | 52m |
Survey of recent nginx features in the 1.11.x and 1.13.x lines (stable 1.12.x), explaining practical use and configuration: the purpose of the --with-compat… | ||
| Avito Data Store: Analytics for a Microservices Architecture | Artem Danilov | 58m |
Avito describes challenges and opportunities for analytics after migrating to a microservices architecture, focusing on new requirements that emerge compared… | ||
| Finding Yourself in the Chaos: Conscious Career Development for IT Professionals | Nataliya Danilenko | 58m |
Describes practices from a large Russian IT company for developing engineers: moving from control (KPIs and bonuses) to trust, adopting mindfulness as a… | ||
| Four Cases of Pair Programming | Dmitriy Simonov | 53m |
The talk examines four real paired-programming (or close-collaboration) cases: a single code owner with ten junior developers, a “tomorrow team” that keeps… | ||
| The Art of Prediction | Andrey Stepanov | 55m |
The talk examines why software time estimates are often wrong and how cognitive biases lead developers to substitute simpler questions (e.g., “how long to… | ||
| How to Kill the Technician in a Manager | Aleksandr Trofimov | 54m |
The talk examines why top engineers promoted to line managers or team leads struggle when their job shifts from coding to people and cross-team coordination… | ||
| Why It Won't Be Found! | Andrey Aksenov | 01h 02m |
Talk examines practical ways to improve search quality, focusing on evaluating and measuring search relevance beyond simplistic thumb-rule checks. It covers… | ||
| ElasticSearch and Heka: How We Learned to Sieve an Elephant Through a Sieve | Adel Sachkov | 54m |
Description of a real-time log collection system built at Yandex.Money that uses ElasticSearch and Heka and currently processes about 2 billion messages per… | ||
| How We Maintain 100 Different Client Versions at Badoo | Yaroslav Golub | 51m |
Badoo runs many projects from a single codebase across iPhone, Android, Web where each platform has a dozen+ concurrent client versions. The talk explains how… | ||
| Life After Sharding | Mikhail Kurmaev | 44m |
Talk discusses post-sharding challenges at Badoo: after physically splitting data, logical relationships remain and data integrity must be maintained. It… | ||
| Why ivi Switched from Sphinx to Elasticsearch | Evgeniy Rossinskiy | 52m |
The talk explains why the streaming service ivi migrated its search backend from Sphinx to Elasticsearch. It describes using the search engine as the core of… | ||
| Quality Control for High-Load Systems | Andrey Drozdov | 47m |
Covers how to do effective load testing for high-throughput services: which metrics to measure, how to interpret results correctly, and how to find bugs that… | ||
| How I Was a Team Lead, and Now — Head of Direction | Vitaliy Sharovatov | 59m |
Describes personal experience transitioning from team lead to head of a department, and practical guidance on selecting and developing a new team lead. Focuses… | ||
| How to Rewrite VKontakte's Private Messaging Database from Scratch | Dmitriy Egorov | 57m |
Describes rewriting VKontakte’s custom personal-messages database engine from scratch because the old engine hindered new features and required massive server… | ||
| How to Write a Service Whose Maintenance Doesn't Turn into Hell | A.Reznikov, Y.Munkyn | 43m |
Covers developing a service with operations in mind for large-scale systems (about 1,500 machines and millions of users). Explains practical logging practices… | ||
| Death by Sale | Anatoliy Plaskovskiy | 49m |
Yandex.Money needed to enable card acquiring for many online stores during Black Friday, which required increasing card authorization throughput by 1,000×… | ||
| How to Build a Good Performance Review: Badoo's Experience | Aleksey Rybak | 01h 00m |
A Badoo engineering team describes how they built an effective performance-review process for software engineers. The talk covers problems with common… | ||
| Blockchain: Lego for the Curious | Aleksandr Borgardt | 46m |
Explains fundamental properties and features of blockchain technology by breaking it into layered abstractions and describing what happens at each lower level… | ||
| Reconstructing YouTube.com — Rewriting the World's Second Most Popular Site | Mikhail Sychev | 01h 02m |
Rebuilding YouTube.com focuses on modernizing the site and developer ecosystem while keeping or improving performance (e.g., maintaining sub-2-second average… | ||
| Resistance in Team Work | Aleksandr Ziza | 50m |
How to handle resistance when a developer becomes a manager of former peers: the talk discusses the challenge of balancing friendly relationships with the need… | ||
| Effective Use of ClickHouse | Aleksey Milovidov | 59m |
Covers practical experience using the open-source columnar database ClickHouse in production, naming ClickHouse as the technology. Describes common usage… | ||
| Voice Assistant Alice | Anatoliy Starostin, Nikolay Lyubimov | 53m |
Overview of Yandex’s voice assistant Alice: its architecture, key technologies (speech recognition, natural language understanding, dialog management)… | ||
| Lua in High-Load Telecom Systems | Dmitriy Borisov | 43m |
Describes using Lua to design and implement high-load telecom systems that meet operators' requirements: scalability, fault tolerance, load balancing, and… | ||
| The Sysadmin's Gentleman’s Kit | Anton Turetskiy | 48m |
Practical examples of sysadmin workflows and production practices used at Badoo, showing how to automate resource-heavy operations to make them simple. Covers… | ||
| Metrics are Not Enough: Monitoring Apache Kafka | Gwen Shapira | 51m |
Explains which concrete signals matter for Kafka operations and why: monitor under-replicated partitions as the primary early failure indicator, track the JMX… | ||
| Monitoring a Cloud CI System: The Jenkins Example | Aleksandr Akbashev | 46m |
Describes building a monitoring system for a large Jenkins-based CI environment with four Jenkins masters (the largest handling over 100,000 builds/runs per… | ||
| Zabbix: Recipes for High-Performance Monitoring | Aleksey Vladyshev | 58m |
Explains principles for building high-performance monitoring using Zabbix, including which Zabbix tools and components to use. Presents benchmark analyses of… | ||
| Rethinking the Approach to Infrastructure as Code | Evgeniy Piven | 39m |
Describes refactoring and modularizing a large legacy Puppet codebase used across many projects, introducing testing for infrastructure code to reduce risk and… | ||
| SOA: Building Our Own Service Mesh | Ivan Kruglov | 59m |
Covers building a custom service mesh for microservice infrastructures, explaining the problems it solves (reliable, secure inter-service communication) and… | ||
| Cheaper, More Reliable, Simpler | Aleksandr Khristoforov | 53m |
Describes building a large-scale video storage and delivery system that handles tens of petabytes and hundreds of gigabits/sec of throughput. Covers the… | ||
| Implementing Our Own Protocol on Top of UDP | Aleksandr Tobol | 01h 00m |
Describes building a custom low-latency streaming protocol on top of UDP for a fault-tolerant streaming service that can survive a datacenter loss and serve… | ||
| Terabits Aren't That Scary | Vyacheslav Olkhovchenkov | 48m |
Shows how a high-load video service was scaled up to a 1-terabit capacity, comparing expectations to real-world results. Describes server-side architecture… | ||
| The Evolution of Databases at Dropbox | Slava Bakhmutov | 01h 00m |
History of Dropbox’s database infrastructure evolution from a single global MySQL deployment to about 6,000 shards. Covers practical engineering: per-type… | ||
| Allow Me to Introduce: MariaDB 10.2 and 10.3 | Sergey Golubchik | 50m |
Overview of MariaDB development and roadmap around versions 10.2 and upcoming 10.3: covers growth in the user base, status of 10.2 as the current stable… | ||
| How and Why to Create an NginX Module — Theory, Practice, Profit. Part 2 | Vasiliy Soshnikov | 51m |
A follow-up talk on writing Nginx modules focuses on unanswered questions from an earlier presentation, specifically covering how and why to implement upstream… | ||
| Reactive Microservices with Apache Kafka | Denis Ivanov | 56m |
Explains using Apache Kafka as the backbone for reactive microservices, describing Kafka’s distributed log abstraction and how it can serve as a… | ||
| Microservice Frontend | Viacheslav Slynko | 53m |
The talk explains how the real-estate site CIAN is moving its frontend from a monolith to a microservice architecture. It describes frontend-specific scaling… | ||
| Introduction to Blockchain and Consensus Algorithms | Filipp Filippak | 55m |
An introduction to how classical blockchains work, covering transactions, blocks, digital signatures, fork causes, and why mining doesn't always require GPUs… | ||
| Metadata for the Cluster: A Key-Value Heroes Race | Ruslan Ragimov, Svetlana Lazareva | 48m |
Compares key-value databases as candidates for storing metadata in a high-performance storage cluster, examining whether popular key-value systems can handle… | ||
| How to Build a Data Platform from Scratch | Pavel Tarasov | 51m |
How to build a data platform from scratch: explains why a company needs centralized data collection (for BI, machine learning integration, or customer data… | ||
| Lambda Architecture for Real-Time Analytics — Risks and Benefits | Nikolay Golov | 52m |
Examines using the Lambda architecture for real-time Big Data analytics, focusing on delivering analytics that lag by seconds or milliseconds rather than hours… | ||
| CTO: Do Dreams Come True? | Roman Ivliev | 59m |
A reflective talk examining the role of engineering leaders and executives in technical organizations, questioning the belief that only engineers do the real… | ||
| How to Communicate with Subordinates and Stay Friends | Aleksandr Ziza, Virna Shtern | 56m |
Advice for engineers who become managers of their former colleagues and friends, explaining how to balance maintaining personal relationships while taking… | ||
| How to Become a Team Lead | Andrey Ryzhkin | 57m |
Defines what a team lead (timlid) is and how that role varies across companies, based on the presenter’s experience building a team-lead program. Describes the… | ||
| The Team as a High-Load System | Anton Potapov | 56m |
Maps common software-development practices to team management by treating a team as a high-load system. Uses developer tools and approaches — such as… | ||
| Experience Developing, Debugging, and Deploying a Hot-Swap Trading System | S.Kostanbaev | 51m |
Presentation explains why and when a hot-standby (active-passive) backup for a trading system is needed, and why Moscow Exchange chose to implement one. It… | ||
| Distributed Fault-Tolerant Financial Transactions Service | Aleksey Burylov | 54m |
A description of building a fault-tolerant, distributed, and consistent financial transaction service capable of 100,000 transactions per second. Covers why… | ||
| Payment System Architecture: Almost Enterprise | Filipp Delgyado | 54m |
A presentation describes the architecture of a modern payment system built from scratch over three years (two in production), focusing on designing reliability… | ||
| NoSQL Best Practices for PostgreSQL | Dmitriy Dolgov | 53m |
Compares Postgres jsonb vs Mongo/BSON with 1M rows benchmark on AWS (Postgres 9.6, Mongo 3.4, custom fork of rnb-mark); shows jsonb throughput matches document… | ||
| Nonrelational SQL | Andrey Nikolaenko | 44m |
Discussion of how SQL has been adapted to query non-relational data sources, surveying SQL-like dialects (HiveQL, AQL, CQL), SQL-on-Hadoop systems and engines… | ||
| Cassandra for Metadata Storage: Successes and Failures | Andrey Smirnov | 57m |
Describes using Apache Cassandra to store metadata for an exabyte-scale S3-compatible object storage. Covers challenges of metadata (concurrent writes/deletes… | ||
| WAMP[-proto] as the Basis for Composite SOA Applications | Konstantin Burkalev | 47m |
WAMP (Web Application Messaging Protocol) is presented as an open protocol that implements both RPC and Publish/Subscribe patterns. The talk argues WAMP is a… | ||
| How packagecloud.io works | Julio Capote | 29m |
packagecloud automatically extracts package metadata, atomically generates repository indexes and GPG‑signs each repo key held in AWS KMS, stores packages on… | ||
| Building Portable Infrastructure for New Relic | Mikhail Panchenko | 53m |
Describes building a portable New Relic infrastructure region to serve European (GDPR-restricted) customers. Covers creating a standalone region for New… | ||
| The Long Transaction — When Size Matters | Mikhail Balayan | 52m |
Examines why long-running transactions degrade PostgreSQL performance by analyzing one real process example. Covers transaction isolation levels, row… | ||
| Distributed Postgres with Citus | Will Leinweber | 44m |
Citus turns Postgres into a sharded, multi node database by adding a Citus extension to standard Postgres servers and a coordinator that rewrites queries into… | ||
| Keeping Up with the Times, or How to Upgrade PostgreSQL | Andrey Salnikov | 49m |
How to perform PostgreSQL version upgrades in production, covering the challenges of upgrading a critical primary data store. The talk discusses practical… | ||
| Logical Replication and Avito | K.Evteev, M.Tyurin, S.Burladyan | 01h 50m |
Avito uses PostgreSQL logical replication to handle growing data volume and query load. The talk describes using logical replication for scaling and load… | ||
| Debugging and Troubleshooting PostgreSQL Streaming Replication | Aleksey Lesovskiy | 49m |
Covers PostgreSQL streaming replication (introduced 2010): why installations use it, common operational problems that arise, and how to diagnose and fix them… | ||
| pg_shardman: Sharding in PostgreSQL | A.Sher, S.Kelvich | 47m |
An experimental approach to sharding PostgreSQL using the foreign data wrapper (FDW) mechanism, the pg_pathman extension for partitioning, and PostgreSQL 10's… | ||
| Scaling DNS | Artem Gavrichenkov | 40m |
The talk examines DNS's long history and practical scalability issues that have accumulated since the original RFCs (882/883) from the 1980s. It discusses… | ||
| One-cloud — Data Center Management System at Odnoklassniki | Oleg Anastasev | 56m |
One-Cloud is a data-center management system used at Odnoklassniki to control over eight thousand physical servers across multiple data centers. It organizes… | ||
| Disaggregation and Openness in Ethernet Networks | A.Petrovskiy, D.Krikov | 58m |
Describes disaggregation in Ethernet networking: separating network device hardware from its operating system and applications so different OSes and network… | ||
| Building a BigData Platform for Russian Post (FGUP Pochta Rossii) | Andrey Bashchenko | 54m |
A Big Data platform built for the Russian Post (FGUP Pochta Rossii) and how it supports the organization’s transformation. Covers the platform’s architecture… | ||
| Schematisation in the Practice of Solving Managerial Problems | Pavel Mrdulyash | 59m |
Schematicization is presented as a method for solving everyday management problems by externalizing conflicting items from working memory onto paper, creating… | ||
| Everything a Team Lead Must Know About Hiring and Firing | Stepan Ovchinnikov | 57m |
Covers what a tech lead needs to know about hiring and firing developers, focusing on developer motivation (both immediate reasons to join a project and… | ||
| Scaling the Network for CinemaVR Attractions | Andrey Tatarinov | 36m |
A presentation about scaling CinemaVR, a network of VR arcade locations deployed in malls and cinemas that grew from 10 to 25 locations within months and… | ||
| Applying Blockchain in RTB | Aleksandr Shvets | 59m |
The talk examines using blockchain in real-time bidding (RTB) advertising systems, focusing on scalability and practical integration. It compares on-chain vs… | ||
| Looking for the Holy Grail of Mobile Web Performance | Adam Bar | 51m |
Крупнейшая профессиональная конференция для разработчиков высоконагруженных систем Saint HighLoad++ 2026
Подробнее: https://clck.ru/3QZHTb
Июнь… | ||
| High speed load balancing from the Linux Kernel | Laura Garcia Liebana | 27m |
ZEVENET builds a Linux-kernel layer‑4 core using Netfilter via nftables, conntrack helpers and a custom nft module set (random hashing, DNAT/SNAT models) and… | ||
| Linux kernel TLS and HTTPS | Aleksandr Krizhanovskiy | 46m |
The talk explains why TLS is being integrated into the Linux kernel and compares implementation approaches from Facebook/RedHat, Mellanox, and the Tempesta FW… | ||
| New Data Replication Technologies in PostgreSQL | Aleksandr Alekseev | 52m |
Overview of PostgreSQL replication and auto-failover capabilities, including new features introduced in PostgreSQL 10. Covers types of replication and the… | ||
| PostgreSQL Configuration for Humans | Álvaro Hernández | 52m |
PostgreSQL ships with ~270 tunable parameters and a 650‑line sample postgresql.conf, and changing about 15 settings on an 8GB/2CPU EC2 instance raises… | ||
| Inexpensive Datamasking for MySQL with ProxySQL | René Cannaò | 51m |
ProxySQL adds RE2 and PCRE regex support and uses table-driven query rules to rewrite or block queries on the fly for cheap MySQL datamasking. Rules keep first… | ||
| Experience Developing a Firewall Module for MySQL | O.Broslavskiy, D.Kolegov | 32m |
A team describes building a MySQL firewall module using the ANTLR parser generator and Kotlin. They cover when ANTLR is appropriate, how to design an ANTLR… | ||
| ProxySQL Use Case Scenarios | Alkin Tezuysal | 55m |
Protocol-aware MySQL proxy that actually implements connection pooling/multiplexing, on-the-fly read/write split, query rewrite rules, per-query caching and… | ||
| MySQL Replication — Advanced Features | Petr Zaytsev | 49m |
Covers MySQL 5.7 replication features and practical tradeoffs: GTID simplifies automated failover and precise transaction tracking but needs… | ||
| Internal Open Source | Pavel Solomin | 56m |
Describes how Sberbank Online scaled its mobile app development by adopting an "internal open-source" process: opening feature development to many internal… | ||
| All Power to the Councils, or Committee-Based Development in Testing | Nina Shcheglova | 51m |
A manager-level case study of growing a small QA department into a large testing organization inside Lazada’s ecommerce project. It covers challenges of… | ||
| Preparing a Test Environment, or How Many Test Instances Do You Need | Aleksandr Dubrovin | 39m |
Describes how to set up test environments for large, evolving projects where dozens of features are developed and tested in parallel. Covers challenges beyond… | ||
| Goth2Boss: Breakdowns and Withdrawals When Transitioning from Engineer to Team Lead | Artem Kalichkin | 01h 09m |
A senior engineering manager with 11 years’ experience explains how to transition engineers into team leads: how to spot leadership potential in strong… | ||
| Key transparency: Blockchain meets NoiseSocket | Aleksey Ermishkin | 01h 01m |
Describes using a sparse Merkle tree to store identifier→public‑key bindings so clients get strict proof of absence as well as presence, anchoring each epoch… | ||
| Pixel Shaders for Web Developers | Denis Radin | 53m |
Covers pixel shaders for web developers: explains why pixel shaders became important, how they work, and how they can be used in ordinary web applications… | ||
| Machine Learning in E‑commerce | Aleksandr Serbul | 53m |
Covers building and operating large-scale predictive analytics for an e-commerce/CRM product (Bitrix24): designing, training and tuning Spark MLlib models in… | ||
| Attribute Detection and Visual Search in UGC Clothing Photos | Dmitriy Solovev | 40m |
Describes a system for detecting clothing attributes and performing visual search over user-generated clothing photos. Covers using computer vision and machine… | ||
| Face Recognition: From Scratch To Hatch | Eduard Tyantov | 52m |
Mail.ru Group builds a production face system: detector + recognizer with 5-landmark alignment, trained on MS‑Celeb plus a cleaned VK crawl (reduced to ~46k… | ||
| Calculating by IP: How to Fight Spam on a Social Network | Mikhail Ovchinnikov | 59m |
Badoo’s anti-spam system for its dating social network: how to detect and block spammers, scammers and adult-content spreaders by classifying billions of user… | ||
| Anomaly Detection in Time Series | Pavel Filonov | 45m |
Covers methods for detecting anomalies in time series data as dimensions and data rates grow. Discusses algorithmic approaches and practical techniques for… | ||
| Pattern Search (Data Mining: Pattern Discovery) | Konstantin Ignatov | 47m |
Discusses methods for discovering repeating patterns in datasets—such as frequent itemsets, substrings, subsequences, subgraphs, image elements, or video… | ||
| 5 boring performance improvements in MongoDB 3.4 that matter | Henrik Ingo | 51m |
MongoDB 3.4 speeds real tasks: initial sync builds indexes in parallel so a ~660GB test drops from about a week to under 4 hours; balancer moves off mongos to… | ||
| Balancing at Wire Speed | Andrey Domas, Vyacheslav Morozov | 50m |
Describes how Odnoklassniki’s traffic load balancer handles high-rate network attacks (especially SYN-floods) and why balancer throughput becomes the critical… | ||
| Traffic Interception — Myths and Reality | Evgeniy Uskov | 51m |
Examines real-world Internet traffic interception incidents (like Google's misrouting in Japan) that caused widespread service outages; analyzes causes such as… | ||
| Measuring performance variabillity of EC2 | Henrik Ingo | 48m |
MongoDB performance tests run 25 times each to measure variability and the big findings are specific: disk noise from ephemeral SSDs is worse than EBS with… | ||
| And Then the Clouds Will Surely Start to Dance! | Aleksey Sushkov | 33m |
Covers virtualization, NFV and SDN as foundations for telco clouds and cloud platforms, explaining basic virtualization concepts, how NFV and SDN are used by… | ||
| Prometheus as a Time Series Database | Timur Nurutdinov | 46m |
An engineer from Lamoda explains using Prometheus as a time-series database for monitoring: they collect millions of metrics and find Prometheus handles heavy… | ||
| How We Got Druid to Work at Odnoklassniki | Yuriy Nevinitsin | 45m |
Describes replacing a 50 TB MSSQL-based analytics system with Apache Druid to fix slow queries, instability (up to 30-minute lag), and high Microsoft licensing… | ||
| Infrastructure as code might be literally impossible | Joe Domato | 45m |
Argues that reproducible infrastructure fails because core tools break in concrete ways: hardware atomics on an Opteron revision fail so MySQL detects and… | ||
| Climbing out from the digital mud with unikernels | Ian Eyberg | 43m |
Claims that tiny VMs stop common attacks by default and remove shells from servers are central: no shell, one process per VM limits RCE and shrinks attack… | ||
| Operating Container-Based Infrastructures | Nikolay Sivko | 52m |
Covers operational challenges of running container-based infrastructures (Docker, Kubernetes, CI/CD, service discovery) in production under load. Focuses on… | ||
| Flexible Data Schema in MySQL (JSON) | Aleksandr Rubin | 51m |
Covers storing IoT sensor data in MySQL using JSON to handle changing schemas. Compares implementation options and JSON features in MySQL 5.7 and MySQL 8.0 for… | ||
| Why are we excited about MySQL 8? | Petr Zaytsev | 53m |
MySQL 8 introduces a native data dictionary so DDL becomes atomic and schema lookups are hundreds to thousands of times faster, invisible and descending… | ||
| Protecting the Web at a scale using consul and Elk | Valentin Chernozemski | 49m |
Architecure uses ELK (Filebeat → Logstash → Elasticsearch → Kibana) plus Consul + consul-template + ipset to detect, score and distribute blacklists for… | ||
| Fraud in mobile applications: how to define and detect | Vadim Antonyuk | 36m |
Two practical detectors for mobile-app RTB fraud: a background ad detection metric that measures each app–device pair’s maximum silent interval (t_max) and… | ||
| CSRF Vulnerabilities Are Still Relevant | Mikhail Egorov | 36m |
The talk explains why Cross-Site Request Forgery (CSRF) remains relevant for web application security, describing current defenses such as anti-CSRF tokens and… | ||
| Optimizations for Yandex Search Results | Ivan Khvatov, Sergey Lyadzhin | 45m |
Yandex Search Results Page (SERP) is a complex, multifunctional web app with strict page-load performance needs, and mobile is slower than desktop due to phone… | ||
| The State of State | Aleksey Okhrimenko | 57m |
Compares approaches to managing application state, naming and evaluating Services (traditional service-based design), CQS/CQRS, Redux (stateless unidirectional… | ||
| Rethinking Relational Database CRUD Design Patterns Since We Live in a Big Data World | Chris Bohn | 47m |
Argument: columnar analytics work best if production systems stream every change as immutable events and analytics store keeps full history while a snapshot is… | ||
| Cloud and Shadow Recognition in Satellite Imagery | Anatoliy Filin | 01h 00m |
Describes detection of clouds and cloud shadows in Landsat-8 satellite imagery, comparing traditional methods like Fmask (which need careful parameter tuning… | ||
| Stored Procedures in NoSQL DBMSs: The Tarantool Example | Denis Linnik | 47m |
The talk examines using stored procedures in NoSQL databases to speed up data processing, using Tarantool as the example. It treats Tarantool as a full… | ||
| Synchronizing Data from PgSQL to Tarantool | Veniamin Gvozdikov | 23m |
Describes synchronizing data from PostgreSQL (PgSQL) into Tarantool to build a fast cache in front of a primary database and reduce load on PostgreSQL… | ||
| Dataplane networking acceleration with OpenDataplane | Maksim Uvarov | 38m |
OpenDataPlane provides a portable C API and a reference C library that runs in userspace and maps to hardware features (queues, classifier, scheduler, crypto)… | ||
| How We Made a Game for HL++ with Voxel Graphics | Aleksandr Khaerov | 56m |
Ingram Micro Cloud built a browser-based multiplayer voxel-graphics game called TheURBN (urbn.odn.pw) for their HighLoad++ booth where players capture… | ||
| BigMail: How We Built a DataLake at Russian Post | Aleksey Vovchenko | 01h 00m |
Describes building a Big Data infrastructure (Hadoop-based Data Lake) for Russian Post: the chosen technologies (Hadoop ecosystem), reasons for those choices… | ||
| libfpta — Outrunning SQLite and Tarantool | Leonid Yurev | 54m |
Libfpta is a new C/C++ storage engine called “Positive Tables” designed as a shared in-memory store for tabular and key–value data. It targets very high… | ||
| Fault-Tolerant Architecture of a Bank's Front-End System | R.Shekhovtsov, A.Gromatchikov | 47m |
A talk about building a fault-tolerant front-end system for a bank, explaining how a bank’s high-load front-end differs from a typical web project and where it… | ||
| How to Build a Cluster to Run Hundreds of Thousands of High-CPU/High-MEM Jobs | Aleksandr Kirillov | 29m |
A cloud CI service that runs customers' tests needed an architecture to quickly scale a cluster up and down for hundreds of thousands of high-CPU/high-memory… | ||
| Building a WebScale Search Engine with a Data Driven Architecture | Tarek Sheasha | 34m |
Built a homegrown Sherlock reservation search serving Booking.com’s >1 billion bookings at ~2,000 req/s with a <50 ms goal, replacing off‑the‑shelf engines… | ||
| Honest Performance Testing | Dmitriy Pivovarov | 35m |
The talk explains why standard statistical metrics (means, medians) are misleading for performance testing of applications and argues that average response… | ||
| Challenges of Performance Testing | Andrey Akinshin | 50m |
Discussion of challenges in writing and integrating performance tests for applications, describing common problems that arise when creating performance tests… | ||
| DevOps Transformation — Breaking Down Walls | Konstantin Pakulin | 45m |
Describes transforming DevOps practices in a large, 15-year-old product with millions of lines of code and many external partners where feature delivery to… | ||
| DevOps Transformation at Alfa-Bank | Anton Isanin | 52m |
Alfa-Bank’s DevOps transformation is presented as a way to accelerate product development and market response in a competitive banking sector. The talk… | ||
| Speeding Up ASP.NET Core | Ilya Verbitskiy | 57m |
Covers performance tuning for ASP.NET Core, the cross-platform web framework built on .NET Core. Describes how to develop on macOS and deploy ASP.NET Core apps… | ||
| Implementing an Order Tracking System | Ravil Garafutdinov | 41m |
A practical case study of building an order-tracking system for Yakitoria: collecting courier GPS data, initial filtering, map-matching to streets, and… | ||
| How to Test Performance of Hyperconverged Systems? | Roman Khmelevskiy | 26m |
Overview of performance testing for hyperconverged infrastructure using Nutanix X-Ray, an open automated tool that runs realistic workloads and fault-injection… | ||
| Mobile Application Performance Analysis | Nikolay Moskalenko | 55m |
Discusses measuring mobile app performance, arguing lab tests on emulators or a few developer devices don’t reflect real-world behaviour. Describes analyzing… | ||
| 100,500 Ways to Cache in Oracle Database | Aleksandr Tokarev | 42m |
Examines why result caching exists in DBMSs and which systems include it, then compares multiple ways to cache SQL query results and database-stored… | ||
| Apache Ignite Persistence | Artem Shitov | 59m |
Apache Ignite now adds native disk storage to the in-memory platform so SQL queries and distributed computations work over the full dataset (not just cached… | ||
| Database Monitoring Mechanisms: An Inside Look | Dmitriy Emanov | 48m |
Covers practical monitoring mechanisms inside database systems for high-concurrency, high-load environments, focusing on tools and DBMS features that DBAs and… | ||
| Java and Linux — Operational Peculiarities | Aleksey Ragozin | 57m |
Covers practical operational nuances of running Java on Linux and inside Docker, explaining how JVM behavior interacts with Linux resource management and how… | ||
| How We Learned to Fix Planes Midflight | Evgeniy Kolomeets | 33m |
Presentation describes implementing Userspace Live Patching: applying safe, on-the-fly code changes inside running user processes (not kernel), while… | ||
| Nuances of a Tracing JIT Compiler | Anton Soldatov | 50m |
Describes practical experience using Lua and LuaJIT for business logic, explains why the team forked LuaJIT due to architectural limits, and discusses details… | ||
| Multiplayer with Go+Lua | Sergey Lerg | 54m |
An example implementation of a multiplayer system using Go for the server side and Lua on the client side. It discusses using Go’s concurrency (channels) to… | ||
| Using Ubuntu Core for IoT in Production | Renat Galimov | 39m |
Describes using Canonical's Ubuntu Core OS to run Internet of Things production devices, focusing on solving device update problems that previously caused many… | ||
| Building Custom Firmware for an IP Camera in Rust | Maksim Lapshin | 52m |
A team describes building a custom firmware for inexpensive IP cameras using Rust. They explain problems with existing cameras: outdated unpatched software… | ||
| Columnar Table Performance Enhancements Of Greenplum Database | Xiaojian Fan | 39m |
Alibaba extends Greenplum (HybridDB) with block-level min/max metadata and disk sort keys, implemented inside columnar files and generated on-the-fly metadata… | ||
| PostgreSQL worst practices | Ilya Kosmodemyanskiy | 47m |
A rapid-fire catalog of concrete anti-patterns for PostgreSQL with how often they appear and quick demos: testing indexes on tiny dev tables and concluding… | ||
| Optimizing High-Contention Writes in PostgreSQL | I.Panchenko, O.Bartunov | 46m |
The talk examines optimizing PostgreSQL for high-contention write workloads, discussing how standard benchmarks (pgbench) shaped PostgreSQL into a… | ||
| CI/CD Best Practices with Kubernetes and GitLab | Dmitriy Stolyarov | 55m |
The talk describes real-world practice of migrating many customer projects into Kubernetes and building CI/CD pipelines with GitLab. It covers deploying 13… | ||
| Kubernetes Configuration: Tips and Tricks | Mikhail Prokopchuk | 45m |
Practical guidance on running Kubernetes for microservices at Avito, covering hardware and OS tuning (network, CPU), Kubernetes infrastructure component… | ||
| Logging and ranting | Vytis Valentinavičius | 48m |
Company standardizes on a central JSON-over-stdout contract so containers and nginx emit structured logs, routed into an ELK stack; dynamic mapping causes… | ||
| Pitfalls Map in Log Collection and Delivery — Lazada Way | Yuriy Bushmelev | 42m |
Describes Lazada’s end-to-end logging pipeline for microservice architectures, covering how applications should write logs, where to send them, how to deliver… | ||
| Storage Technologies for Large-Scale Projects | Sergey Platonov | 39m |
Compares storage technologies for large projects by explaining differences between distributed, shared, and parallel file systems and when Scale-In… | ||
| I Want to Compress Everything | Andrey Aksenov | 52m |
Overview of data compression techniques and how choosing the right compression scheme can both save disk/memory and sometimes speed up code. Covers compression… | ||
| Incident Management at Badoo | Ilya Ableev | 48m |
Describes incident management practices at Badoo for high-load production services, covering how teams detect, track and control incidents that arise… | ||
| How to Serve a Billion Users and Deliver Terabits of Traffic | Igor Vasilev | 51m |
Examines how to serve a billion users and deliver terabits of traffic by focusing on network engineering for high load. Covers architectural patterns used at… | ||
| DDoS Resilience by Design | Ramil Khantimirov | 52m |
The talk explains the concept of "protectability"—how easy a system can be defended from DDoS—and argues it should be treated as a design parameter. It… | ||
| Aleksey Akulovich at HL++ 2017 | Aleksey Akulovich | 46m |
Presentation covers the HighLoad++ conference event details: it names the professional conference for developers of high-load systems (Saint HighLoad++), gives… | ||
| Beyond WHERE and GROUP BY | Sergey Golubchik | 44m |
Shows which MariaDB SQL features to actually use and what they do: recursive CTEs for graph traversal (friend-chain/BFS example and Fibonacci/date generation)… | ||
| How to Evolve a Component Library Without Breaking It | Artur Udalov | 46m |
A Mail.ru team describes managing a shared UI component library used across projects and the repeated breakages when developers customized it. They explain… | ||
| What to Do Tonight If I Know How Many ++i + ++i Will Be | Andrey Borodin | 35m |
Presents ideas for student projects that apply efficient algorithms and data structures to improve open-source software. Reviews existing code snippets… | ||
| Surviving an Online Game Launch and Dealing with Dead Souls | Andrey Trifanov | 58m |
The talk examines technical challenges during an online game launch, focusing on first-time player experience ('first minutes') and how that design emphasis… | ||
| Neural Networks: Fast GPU Inference with TensorRT | Dmitriy Korobchenko | 43m |
Describes accelerating neural-network inference on GPUs using NVIDIA TensorRT: explains that training frameworks trade performance for flexibility, so deployed… | ||
| Gobblin as an ETL Framework | Ivan Akhlestin | 47m |
Overview of ETL as a class of data-extraction, preparation and delivery tasks in the Big Data ecosystem, explaining where ETL fits among other approaches and… | ||
| Causal Consistency in MongoDB Explained in Detail | Mikhail Tyulenev | 42m |
The talk explains how MongoDB implements causal consistency to avoid anomalies of eventually consistent distributed databases where reads on different nodes… | ||
| Pain-Free Fallback Cache in Scala | Oleg Nizhnikov | 51m |
Describes Tinkoff’s approach to building a resilient fallback cache for Scala backends in microservice architectures so services keep responding when important… | ||
| Roundtable: What the CEO Expects from a Team Lead and What the Team Lead Wants from Their CEO | verno shtern, tatyana pakha rezkaya, konstantin chumachenko, odin iz kochergin | 01h 37m |
A panel discussion about the mismatch between what CEOs expect from team leads and what team leads expect from their CEOs, focusing on roles such as team… | ||
| nothing matches | ||