Issue link: https://read.uberflip.com/i/1476424
Scale You need to develop your applications with scale in mind, especially during growth phases, such as when you enter new markets or expect an increase in the number of users that interact with your applications. There are generally two scenarios to consider when planning for scale: • Short-term traffic spikes: Modern-day applications are bursty by nature. Whether we're talking about websites, APIs, or streaming applications, the demand on your application is likely to fluctuate depending on a variety of different circumstances—often unpredictable in nature. These unanticipated spikes in demand can result in dozens or even hundreds of times the normal load on your application's compute, storage, and networking infrastructure. Hitting a scale limit in your architecture could impact your ability to grow your business. However, with the right architecture, your applications can be built to handle today's needs and the growth you'll experience tomorrow. • Long-term growth: As your business grows over time, you need to be prepared to scale out in new directions, without painting yourself into a corner. While your current infrastructure may meet your current needs, it may not be able to provide the same experience to users in new markets. Similarly, as traffic grows, you may find that costs soar and performance deteriorates due to architectural limitations that inefficiently utilize your servers. Given the right selection, you can alleviate these challenges by having the right infrastructure available to run your applications where and when you need them. To achieve this, long-term growth plans need to consider not only how you can scale your existing architecture but also how you can optimize your costs, performance, reliability, and security as your demand and application types evolve. If you don't plan out your application architecture with the right infrastructure choices to handle both short-term and longer-term scale challenges, the impact can be significant. Exceeding the capacity your architecture was designed for can affect your application, leading to slow performance, unhappy customers, and extra maintenance. Even if your response team prevents your application from going down, identifying the issue and unexpectedly provisioning new resources is costly and time-consuming. 4