Percona Live Open Source Database Conference 2017
2017-04-24..2017-04-27
Santa Clara, United States
| title | author | length |
|---|---|---|
| Database Stalls, From the Ordinary to the Obscure | Preetam Jinka | 35m |
VividCortex detects short database stalls down to 1 second by modeling Little’s Law and watching threads_running (concurrency) against queries per second… | ||
| High Availability in GCE | Carmen Mason, Allan Mason | 47m |
Migration moves MySQL from a Tennessee data center to GCE using MHA for topology and ProxySQL on each app host to route traffic; MaxScale and Orchestrator are… | ||
| Edgestore Multi-tenancy & Isolation | Bogdan Munteanu | 46m |
Edgestore runs at multi‑million QPS serving trillions of objects on thousands of MySQL shards (248 logical shards visible per deployment), and solves… | ||
| Lock, Stock and Backup: Data Guaranteed | Andrew Moore, Jervin Real | 46m |
Recommends a three-part approach: binary log streaming for point‑in‑time recovery, plus mydumper (parallel logical dumps but no streaming) for quick partial… | ||
| #MySQL on Ceph using Rados | Yves Trudeau | 50m |
Prototype implements MySQL storing InnoDB pages on Ceph via RADOS and RBD, using snapshot+clone provisioning to spin slaves that share a base image (demo shows… | ||
| Histograms in MySQL and MariaDB | Sergei Petrunia | 45m |
MariaDB has had histograms since 10.0, stores compact histograms in mysql.column_stats (about 225 bytes) and collects them only via full table scan using… | ||
| Automating database migration - Amazon DMS tutorial | Eran Schitzer | 39m |
AWS Database Migration Service (DMS) plus the Schema Conversion Tool (SCT) automates moving and modernizing databases with concrete results: launched March… | ||
| Multi Active-Active and disaster recovery with MongoDB database center | Xixuan Xixuan | 29m |
Built a cross-region replicator called Lamda with manager, collector, tunnel and receiver that streams oplog entries between clusters, uses a global ID to… | ||
| Automating high availability MongoDB cluster deployment and management | Jonathan Rudenberg | 32m |
Flynn provisions and manages three-node replica set MongoDB clusters with an external state machine and service discovery, using discoverd with Raft to store… | ||
| InfluxDB storage engine internals | Paul Dix | 40m |
Built a bespoke storage engine centered on a TSM tree that writes to a WAL, flushes memory into memory‑mapped TSM files with blocks of up to 1,000 points, and… | ||
| MySQL (NDB) Cluster Best Practices | Johan Andersson | 59m |
MySQL Cluster (NDB) is an in‑memory, sharded ACID store with synchronous replication via a linear two‑phase commit and checkpointing to disk; a betting… | ||
| Operationalizing 19 years of infrastructure via MySQL-as-a-Service | Nathan Northcutt | 26m |
GoDaddy converts 19 years of MySQL sprawl into a MySQL-as-a-Service platform built on Docker containers, an overlay filesystem for separating data from images… | ||
| ClickHouse as Time-Series Storage for Graphite | Dmitry Andreev | 22m |
ClickHouse serves as the time-series store for Graphite with Yandex’s open-source Graphouse acting as the bridge: it replaces Carbon, batches inserts into… | ||
| PhxSQL: a high-availability and strong-consistency MySQL cluster | Ming Chen | 22m |
Ming Chen, Architect of WeChat, Tencent delivers their talk, "PhxSQL: a high-availability and strong-consistency MySQL cluster", on DAY 2 of the Percona Live… | ||
| MongoDB in AWS (MongoDB as a DBaaS) | Jing Wu | 28m |
Automated DBaaS that builds configurable MongoDB clusters on AWS: CloudFormation templates plus Lambda calculate instance counts and preassign member IDs, Auto… | ||
| Using SPIDER for sharding in production | Kayoko GOTO, Kentoku SHIBA | 24m |
SPIDER is a MariaDB/MySQL storage engine that shards by using native table partitioning, supports cross-shard joins with join pushdown and aggregate pushdown… | ||
| Comparing MongoRocks, WiredTiger and mmapv1 for performance and efficiency | Mark Callaghan | 28m |
Comparative benchmarks on MongoDB 3.4 show MongoRocks (RocksDB engine) dramatically reduces space and write work: databases shrink to ~37GB vs 49GB for… | ||
| Do's and Dont's of the modern hybrid mongoDB MySQL environment | Jonathan Wage, Rick Vasquez | 15m |
Recommendation: choose mongoDB for document workloads and MySQL for highly relational data, test MyRocks or InnoDB and MySQL partitioning for high write OLTP… | ||
| Performance Analysis of NVMe SSDs and their Implication on Real World Databases | Qiumin Xu | 25m |
Hardware and software measurements use fio, blktrace and driver instrumentation on a Samsung XS1715 NVMe, a Samsung A43 Pro SATA SSD and a 15K HDD to quantify… | ||
| sysbench 1.0: teaching old dog new tricks | Alexey Kopytov | 34m |
sysbench 1.0 (formal release Feb 2017 after a decade of informal branches) rewrites the Lua workload layer to use LuaJIT FFI and a bundled RNG, measures client… | ||
| Experiences using gh-ost in a multi-tier topology | Valerie Parham-Thompson, Ivan Groenewold | 16m |
gh-ost runs in multi‑tier setups by reading binlogs (requires ROW format and sometimes log_slave_updates) and can run against masters or slaves to offload… | ||
| How to Build a State Machine on MongoDB and Redis | Niki Castle | 24m |
CRM flow engine Canvas pairs MongoDB for single-write user state with Redis for coordination and caching, using Sidekiq/sidekick jobs. Key techniques: a Redis… | ||
| MyFlashSQL: Flash is more than faster-harddisk | Sang-Won Lee | 27m |
MyFlashSQL modifies the MySQL engine and shows concrete gains: changing buffer page size from 16KB to 4KB yields ~2.5x throughput on TPC-C/LinkBench; reworking… | ||
| Demystifying Benchmarks: How to Use Them To Better Evaluate Databases | Peter Friedenbach | 44m |
Tracks TPC history and shows why published scores no longer suffice, then demonstrates practical tool use: YCSB and sysbench/CISbench to drive workloads, iperf… | ||
| Current state of MongoDB Backup options | David Murphy | 26m |
Covers practical backup behavior and tradeoffs: mongodump goes to secondary via --host but mongos hits primary so logical dumps are “squishy,” restores… | ||
| MongoDB size management | Steven Wang | 20m |
Shows concrete steps to cut MongoDB footprint and what actually worked: migrate to WiredTiger with Snappy drops some collections from ~1 TB to ~100 GB… | ||
| Reusable Distributed Data Infrastructure with Dynomite | Ioannis Papapanagiotou | 21m |
Dynomite converts single-node stores into a peer‑to‑peer, master‑master layer that adds sharding, replication and multi‑DC failover to Redis or RocksDB, with… | ||
| How to improve MySQL performance even with Handler Socket with replication to in-memory Tarantool | Alexander Yakovlev | 16m |
Claims a replicator from MySQL to Tarantool that replaces HandlerSocket and yields ~6–7× faster responses by moving hot working set into an in‑memory Tarantool… | ||
| MongoDB Engines: Demystified | Sveta Smirnova | 19m |
MongoDB offers WiredTiger and MMAPv1 plus Percona Server’s MongoRocks and In-Memory, and the talk gives concrete tradeoffs: WiredTiger default since 3.2 with… | ||
| When and How Should you Shard? | David Murphy | 24m |
Explains when sharding actually speeds queries or writes, and how to pick a shard key with measurable evidence: turn the profiler to full, parse logs with… | ||
| Database management made simple - Amazon RDS (Part 1 and 2) | Darin Briskman | 01h 05m |
Amazon RDS runs managed MySQL/MariaDB with daily full backups plus continuous binary logs for point‑in‑time restores up to 60 days, automatic failover that… | ||
| Introducing ScaleDB, the New Clustered, Distributed Storage Engine for MySQL (Part 1) | Ivan Zoratti | 58m |
ScaleDB is a clustered storage engine that plugs into MariaDB and runs as separate storage nodes (cas) + lock manager (slm) with volumes mirrored for… | ||
| Percona Server for MongoDB: Features Overview | Jon Tobin | 25m |
Percona Server for MongoDB adds two storage engines (Percona Memory Engine, MongoRocks (RocksDB-based)), audit, external authentication, HotBackup for… | ||
| MongoDB Security: Making Things Secure by Default | Adamo Tonete | 15m |
MongoDB deployments receive concrete steps: enable authentication (or bootstrap a first admin user before turning it on), use a shared keyfile for internal… | ||
| Optimizing MySQL without SQL or touching my.cnf | Maxim Bublis, Peter Boros | 43m |
Dropbox reworks MySQL builds and OS settings to squeeze real gains: profile‑guided optimization on GCC/Clang and link‑time and native‑arch builds yield up to… | ||
| MyRocks: best practice at Alibaba - MySQL Storage Engine | jiayi wang, dengcheng he | 27m |
Alibaba tests MyRocks (RocksDB) against AliSQL/InnoDB and deploys it to over a hundred production databases, cutting storage to 1/7 or typically 1/3–1/8 of… | ||
| Rolling out a Database-as-a-Service layer at Ghent University | Matthias Crauwels | 25m |
University replaces end‑of‑life physical MySQL boxes with a private VMware DBaaS running Percona Server 5.7, managed by Puppet and offering self‑service… | ||
| Common Table Expressions and Window Functions simple, maintainable, fast queries | Vicentiu-Marian Ciorbaru | 34m |
MariaDB Server 10.2 adds Common Table Expressions and window functions and shows concrete performance gains: rewriting running-account sums with… | ||
| How innovation happens on Amazon Redshift | Eric Ferreira | 42m |
Amazon Redshift is a Postgres‑based MPP data warehouse reworked with columnar storage and per‑column compression, in‑memory zone maps and flexible distribution… | ||
| ToroDB Stampede: open-source BI connector for MongoDB | Álvaro Hernández | 33m |
ToroDB Stampede is an open-source secondary replica that reads MongoDB oplogs, auto-generates relational schema and writes live relational tables to a… | ||
| Deploying #MongoDB on public clouds: Architecture and options | Dharshan Rangegowda | 50m |
ScaleGrid reports running thousands of MongoDB instances on AWS and Azure and gives concrete rules: replica sets allow up to 50 secondaries but only 7 voting… | ||
| Six New Important RocksDB Features And Planned Works | Siying Dong | 51m |
Six concrete additions: a user‑space persistent cache that appends compressed blocks to log files, evicts whole files and maps block IDs in memory; direct I/O… | ||
| Building a scalable time-series database on PostgreSQL | Michael J. Freedman | 51m |
TimescaleDB implements a PostgreSQL extension that exposes a single hypertable while automatically creating right-sized chunks via adaptive time-space… | ||
| Understanding all the Options for Replica-sets and Replication Patterns | Bimal Kharel | 36m |
Explains concrete MongoDB replica set choices: clusters can have up to 50 members but only 7 voting members matter for elections, elections move to a… | ||
| MySQL/MariaDB Parallel Replication: inventory, use cases and limitations | Jean-François Gagné | 50m |
Inventory of MySQL/MariaDB parallel replication, specific mechanics and real Booking.com benchmarks: MariaDB uses domain ID tagging (10.0) and master-side… | ||
| Amazon Aurora: A deep-dive into what's new | Sailesh Krishnamurthy | 55m |
Amazon Aurora separates storage into a scaleout service that writes only redo log records to six copies and commits after a 4-out-of-6 quorum, reducing network… | ||
| #PostgreSQL to #MySQL: A DBA Perspective | Patrick King | 31m |
Yelp runs MySQL 5.6 with statement-based replication and replication trees up to five nodes deep, supporting ~400 engineers and doing ~15 schema changes per… | ||
| OpenTSDB - Time Series Schema on Schemaless NoSQL | Chris Larsen | 44m |
OpenTSDB history and schema decisions explained with concrete tradeoffs: a production HBase deployment handling about 10 million writes per second (peaks ~18M)… | ||
| #MongoDB Index Types, How, When and Where Should They be Used? | Adamo Tonete | 25m |
MongoDB has 12 index types and the talk shows concrete tradeoffs: collection scans on 100,000 documents take about 350 ms on 100k docs without an index but… | ||
| Multi-Site, Multi-Master Done Right | Matthew Lang | 35m |
Continuent’s stack uses a local HA cluster plus an asynchronous cross-site replicator to run multi-site, multi-master MySQL with no application changes… | ||
| MySQL GTID Implementation, Maintenance, and Best Practices - The Short Version | Mark Filipi, Brian Cain | 49m |
GTID adoption is practical and concrete: Percona Server 5.6 uses gtid_deployment_step to enable GTIDs online without full downtime while the conversion… | ||
| #ProxySQL Use Case Scenarios | René Cannaò | 50m |
ProxySQL sits between apps and MySQL and actually reduces connection overhead (examples claim ratios like 100,000 app connections to 100 DB connections via… | ||
| Automatic MySQL schema management with Skeema | Evan Elias | 52m |
Skeema is a CLI that converts a repo of declarative create-table files into safe DDL: it diffs repo vs live MySQL/MariaDB, emits ALTER/CREATE/DROP, and can… | ||
| Reducing Risk Before Upgrading MySQL | Kenny Gryp, Tom De Cooman | 48m |
Process shows exact steps: provision a spare server, create a MySQL 5.7 replica of 5.6, run pt-table-checksum to find inconsistencies (example: a 44,000‑row… | ||
| Galera Cluster Data Consistency | Seppo Jaakola | 53m |
Explains concrete failure modes and fixes: state snapshot transfers via mysqldump, rsync, xtrabackup can seed a bad node; session flags (wsrep_on/sql_log_bin)… | ||
| Vitess beyond YouTube | Robert Navarro | 51m |
Overview of what Vitess actually does: it serves MySQL traffic via MySQL protocol and V3 drivers, implements transactional 2PC and transactional messaging, and… | ||
| Cookbook for Creating INDEXes -- All about Indexing | Rick James | 57m |
Step-by-step recipe for building useful indexes: pick all columns that are column=constant first, then allow one range column (equal then range is optimal)… | ||
| Replication latest features: the community sceptic roundup | Giuseppe Maxia | 51m |
Coverage of GTID, multi‑source, parallel appliers and group replication focuses on concrete behavior and monitoring: MySQL uses UUID GTIDs and leaves GTIDs off… | ||
| Meet Percona XtraDB Cluster 5.7 | Krunal Bauskar | 47m |
Percona XtraDB Cluster 5.7 GA (six months) adds concrete controls and tooling: cluster safe mode blocks non-transactional DDL/DML and missing PK writes… | ||
| MySQL Load Balancers - MaxScale, ProxySQL, HAProxy, MySQL Router & nginx - a close up look | Krzysztof Ksiazek | 48m |
Compares MaxScale, ProxySQL, HAProxy, MySQL Router and nginx with concrete behaviors: MaxScale 2.0 shows CPU contention and 95th-percentile latency spikes (one… | ||
| MariaDB Server 10.2: The Complete Guide | Colin Charles | 53m |
MariaDB 10.2 ships as a release candidate (2017) and continues the fork-from-MySQL‑5.5 path while adding concrete features: SQL window functions and recursive… | ||
| MySQL 5.7, 8.0 and MongoDB: Geospatial Introduction | Alexander Rubin, Michael Benshoof | 35m |
Quick, concrete takeaways: MySQL 5.7 adds InnoDB R‑Tree spatial indexes and ST_ functions which cut a sample query from ~60 seconds and ~33,000 handler ops to… | ||
| EVCache: Lowering Costs for a Low-Latency Cache with RocksDB | Scott Mansfield | 54m |
EVCache moves terabytes of replicated RAM cache onto SSD-backed RocksDB to cut cost while keeping latency low: Netflix runs hundreds of TB and trillions of… | ||
| Time series collection and processing in the cloud: integrating OpenTSDB with Google Cloud Bigtable | Danil Zburivsky, Christos Soulios | 25m |
An open source add‑on links OpenTSDB to Google Cloud Bigtable so OpenTSDB uses Bigtable instead of running a full HBase/Hadoop stack; the team builds an… | ||
| nothing matches | ||