Handling web servers of high traffic sites
Download as PPTX, PDF2 likes6,351 views
This document discusses various techniques for handling high traffic websites, including on-site improvements like minifying CSS and JavaScript, using content delivery networks, gzip compression, and browser caching. It also discusses server-side improvements such as employing caching, setting up a proxy server with load balancing, using a master-slave database setup, optimizing the database server, and avoiding hitting swap memory. The document concludes with case studies of approaches used by Google, Facebook, and YouTube to handle high volumes of traffic.
Handling web servers of high traffic sites
- 1. Handling High Traffic WebsitesASeminarOnHandling High Traffic WebsitesByAshishKumar aka AshfameCP06023 / 06EJECS022
- 10. Browser CacheOn-Site ImprovementsBrowser CacheSERVERblog.ashfame.com
- 11. On-Site ImprovementsReduce Image size (Image format)Use Content Distribution Network (CDN)Gzip CompressionJIT Approach – CSS at top | JS at bottomAvoid CSS expressions
- 12. Server Side ImprovementsEmploy Caching
- 13. Server Side ImprovementsSetup a Proxy ServerDatabase ServerBack End Server (Apache | IIS)Front End Server (Nginx)HTTP Client
- 14. Server Side ImprovementsMaster Slave DB SetupMasterSlave 1Slave IISlave IIIHeavy LoadUsual Load
- 15. Server Side ImprovementsLoad BalancerSERVER1SERVER IISERVER IIISERVER IVSERVER VSERVER VILoad BalancerRequests (Heavy Load)
- 16. Server Side ImprovementsRAID ConfigurationThreaded Model over Processes ModelMySQL (Database Server) Tweaking SettingsMySQL Storage Engine (InnoDB | MyISAM)Avoid hitting Swap (I/O) at all cost
- 17. Your Approach
- 18. Case study - Google
- 19. Case study - Facebook
- 20. Case study - YouTube
- 21. Queries?