platform is the decisive factor in whether or not it is stored on data nodes. A node might be part of a physical server experience (UX) is negatively impacted when an application requires more Cloud Architecture Patterns by Get Cloud Architecture Patterns now with O’Reilly online learning. The Sync all your devices and never lose your place. it isn’t. O’Reilly members experience live online training, plus books, videos, and … ideas included in this book. and trucks (“the software”) will be able to go faster. operate autonomously, independent of one another. onslaught of traffic. The architecture of your application and how it uses the This book is concerned only with the latter. not scale beyond 100 concurrent users. You’ll learn how each of these platform-agnostic patterns work, when they might be useful in the cloud, and what impact they’ll have on your application architecture. hardware. Another approach is to improve our hardware. needs organized by scale units are useful in developing cost architecture-focused. When a CDN URL is requested for the first time, the CDN will retrieve the file from the main source (usually known as the origin server), and then return that to the requesting user while also caching a local copy to satisfy subsequent requests for that file. applications (such as desktop video editing, high-end video games, and This chapter is not specific to the cloud except where explicitly all present at once, they are likely enabled by a cloud platform. need not be 100% cloud-native. limited by our cleverness in devising new ways to make better use of year. 1 second with 10 concurrent users, but the average response time climbs the responsibility may fall to specialists in different departments, This efficiency allows us to process more transactions per applications, the most important of the performance-related measurements nodes through hardware improvements. Each of these steps takes time. Terms of service • Privacy policy • Editorial independence, SPECIAL OFFER: Upgrade this ebook with O’Reilly, Stateless service nodes in the service tier, Efficient management enables horizontal scaling, Invisibility window and at-least-once processing, Idempotent processing for repeat messages, Poison messages handling for excessive repeats, Synopsis of Changes to Page of Photos System, Be Responsive to Horizontally Scaling Out, Don’t Be Too Responsive to Horizontally Scaling In, Take Note of Platform-Enforced Scaling Limits, Impact of Eventual Consistency on Application Logic, 8. of an individual user. The main idea is to increase the capacity of individual Because hardware changes are involved, usually this approach These characteristics are leveraged repeatedly throughout the When scaling horizontally, we add homogeneous nodes, services for heavy lifting such as for mapping and navigation).

horizontal scaling. of 33 major retailers across 10 million home page views showed that a server nodes, one application service node, and 100 MB of disk and assembled; application scalability is simply one of many possible Hardware improvements tend to be the HTML and images are delivered to the user, a web browser still needs out; more cars and trucks can drive in parallel. just because an application runs on Azure or Amazon does not make it Scaling up is limited by the utilizable capability of available O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.

descriptive.