Orange developers day-Application Caching
?Download as PPTX, PDF?
1 like?876 views
This document discusses caching strategies for web applications and ORM caching. It begins with an overview of why caching is important and how it can speed up applications based on Amdahl's law. It then discusses two levels of ORM caching - first level session caching and second level caching using various cache providers like EhCache. The caching strategies covered are read-only, read-write, transactional and nonstrict read-write. Memory, disk, clustered and server caching types are also mentioned.
Orange developers day-Application Caching
- 1. Web Application Caching DB ORM Caching Gamal Shaaban International Centers Orange Labs P&S Cairo Mobile Applications & Software Development
- 2. Agenda ? Why Caching ¨C The Theory of Caching ? ORM Caching ¨C 1st Level of Caching ¨C 2nd Level of Caching ¨C Query Caching
- 3. Why Caching?
- 4. Why Caching How much faster will caching make an application?
- 5. Why Caching It depends on a multitude of factors being: ? how many times a cached piece of data can and is reused by the application. ? the proportion of the response time that is alleviated by caching. ? In applications that are I/O bound, which is most business applications, most of the response time is getting data from a database. Therefore the speed up mostly depends on how much reuse a piece of data gets.
- 6. Why Caching Amdahl's Law & How to calculate entire system speed up. ? Amdahl's law, after Gene Amdahl, is used to find the system speed up from a speed up in part of the system. 1/ ((1 - Proportion Sped Up) + Proportion Sped Up / Speed up)
- 7. Why Caching Speed up from a Web Page Cache Un-cached page time: 2 seconds Cache retrieval time: 2ms Proportion: 100% The expected server side system speedup is: 1 / ((1 - 1) + 1 / 1000) = 1 / (0 + .001) = 1000 times system speedup The to the browser ¡°system¡± speed up is much less
- 8. Why Caching Speed up from a Database Level Cache Un-cached page time: 2 seconds Database time: 1.5 seconds Cache retrieval time: 2ms Proportion: 75% (1.5/2) The expected server side system speedup is: 1 / ((1 - .75) + .75 / (1500/2))) = 1 / (.25 + .75/750) = 3.98 times system speedup The to the browser ¡°system¡± speed up is much less
- 10. ORM Caching
- 11. ORM Caching 1st Level Cache Session Caching: ¨C Enabled by default. ¨C Used transparently during the session. ¨C All objects that was saved or retrieved ? save ? get ? list
- 12. ORM Caching 1st Level Cache Session Caching: ¨C flush() - will sync cache to DB. ¨C clear() - will evict all objects.
- 13. ORM Caching 2nd Level Caching
- 14. ORM Caching 2nd Level Caching
- 15. ORM Caching Cache Providers ? ? ? ? ? ? ? ? EhCache. OSCache. JBoss Cache (AKA Infinispan). WarmCache. Hazelcast. Oracle Coherence. Gigaspaces. GemFire
- 17. ORM Caching Caching Types ? ? ? ? Memory Caching Disk stores Clustered (distributed Caching) Server Caching (RESTful cache server).
- 18. Questions ?
- 19. Thank You
Editor's Notes
- #17: Read-only:?A concurrency strategy suitable for data which never changes. Use it for reference data only.Read-write:?Use this strategy for read-mostly data where it is critical to prevent stale data in concurrent transactions, in the rare case of an update.Transactional:?Again use this strategy for read-mostly data where it is critical to prevent stale data in concurrent transactions, in the rare case of an update.Nonstrict-read-write:?This strategy makes no guarantee of consistency between the cache and the database. Use this strategy if data hardly ever changes and a small likelihood of stale data is not of critical concern.
- #18: GimFire server cache & Spring data integration.