Как понять, что происходит на сервере? / Александр Крижановский (NatSys Lab., Tempesta Technologies)

[Open Infrastructure & Cloud Native Days Korea 2019] 커뮤니티 버전의 OpenStack 과 Ceph를 활용하여 대고객서비스를 구축한 사례를 공유합니다. 유연성을 확보한 기업용 클라우드 서비스 구축 사례와 높은 수준의 보안을 요구하는 거래소 서비스를 구축, 운영한 사례를 소개합니다. 또한 이 프로젝트에 사용된 기술 스택 및 장애 해결사례와 최적화 방안을 소개합니다. 오픈스택은 역시 오픈소스컨설팅입니다. #openstack #ceph #openinfraday #cloudnative #opensourceconsulting

This document provides instructions on how to use Redis, an open source in-memory NoSQL database. Redis can be used when fast responses are needed or for tasks like session storage, job queues, and real-time rankings. It has advantages like speed and atomic operations but disadvantages like memory overhead. The document explains how to install, run, and test Redis, and demonstrates several of its features including lists, sets, hashes, and publish/subscribe. It also provides recommendations on clustering and shows the results of a performance test.

This document discusses how PostgreSQL works with disks and provides recommendations for disk subsystem monitoring, hardware selection, and configuration tuning to optimize performance. It explains that PostgreSQL relies on disk I/O for reading pages, writing the write-ahead log (WAL), and checkpointing. It recommends monitoring disk utilization, IOPS, latency, and I/O wait. The document also provides tips for choosing hardware like SSDs or RAID configurations and configuring the operating system, file systems, and PostgreSQL to improve performance.

This document discusses benchmarks comparing the performance of PostgreSQL and MySQL databases. It describes running the same tests on both databases using pgbench on a 72-core server. Initial attempts using sysbench to synchronize the tests had issues with NULL handling and loading PostgreSQL. The benchmarks therefore used pgbench scripts for point selects, OLTP read-only and read-write workloads. The results showed that PostgreSQL and MySQL performance was very close. The document also discusses potential bottlenecks in PostgreSQL and ways to improve performance, such as using a true in-memory engine, faster snapshots, and asynchronous protocol.

"We're hiring!" How many times have you heard this phrase at a conference? Every database-driven company is hiring and that makes for pretty stiff competition when trying to get a new DBA. Instead of searching for the perfect database administrator from a conference or Linkedin, why not look internally at your organization for system administrators or engineers who may be an equally good fit given the right training. In this talk, I'll explain how the DBAs at Box developed a knowledge-sharing culture around databases and disseminated important learnings to other members of the company. I'll also cover the mentorship process we established to train other members of our Operations team to become rock star DBAs and manage our MySQL and HBase infrastructure at Box.

1. The document describes load testing done on an Odoo implementation to determine system limits and size requirements. Tests were run using Locust and varied the number of requests per second, users, and sessions to identify bottlenecks. 2. Testing showed the initial limit was around 500 requests/second. Further tests found that sessions stored in NFS caused issues, so they were changed to store in PostgreSQL. 3. Load balanced and monolithic architectures were compared, finding the load balanced setup using 4 servers performed similarly to a single server with 40 cores for the monolithic setup.

This slide is a translation of http://www.slideshare.net/akirahiguchi/handlersocket-plugin-for-mysql-4664154

In this session we will discuss selected areas of InnoDB and XtraDB 5.7 internals that are mostly related to buffer pool management, flushing, and the doublewrite buffer, from a performance and scalability point of view.

1. A PostgreSQL database outage occurred at GitLab on January 31st due to a combination of factors including an increase in load, replication lag, and the deletion of the database directory. 2. Lessons learned include monitoring replication, using tools like pg_basebackup properly, and having backups and disaster recovery processes in place. 3. Recommended preventative measures include setting sane configuration values, automated testing of backups, assigning an owner for data durability, and improving documentation.