BUILT TO SCALE NOT JUST DEPLOYED TO SCALE

Building an enterprise-class product requires scalability as a key component from the day the product is envisioned. In general, product managers tend to focus on their MVP with no idea of scalability which could be easily achieved via cloud computing. Based on my experiences with many products over the years I believe there are two aspects to any SME product:

● Design for the best user experience
● Scale for the user experience

Keep in mind building a scalable enterprise-class product is a process that when we build a product, the objective is to build it to perfection. Until you deploy it to the public you won’t have the volume of real-time data or consumers actually using the product. End-users may interact with the product differently than what you may have expected or they may request additional features, this does happen often.

Scaling a product once it is live is very important for any product to succeed.

ARCHITECTS HAVE TWO OPTIONS TO SCALE THE PRODUCT IN CLOUD COMPUTING:

1. Scale Up (Vertically) – Product is deployed on over-priced hardware/infrastructure
2. Scale Out (Horizontally) – Product is deployed on multiple computers. If built properly it can provide scalability at a very low cost. Product design and development will be more complicated though so you need a specialized skill-set to design and build your product.

Most businesses develop products on the cloud with a pay-as-you-go model. The main advantage is that the cloud computing based architecture ensures scalability and decreases other complexities. Other terms used are: Microservices based architecture and development. Microservices are a shift from traditional software development architecture that has gained popularity in the last 5-6 years. Microservices architecture has been used by some major consumer-facing technology companies such as Amazon, Twitter, and PayPal.

In general, a scalable product must have a well-defined database architecture and data strategy.

THERE ARE A FEW KEY AREAS THAT I WOULD RECOMMEND FOCUSING ON, FROM A SCALABILITY STANDPOINT:

● Query performance and optimization
● Data caching
● Replica servers
● Database shards

AS FAR AS SERVERS ARE CONCERNED, BELOW ARE THE AREAS THAT ALSO NEED TO BE WELL THOUGHT OF:

● Proven technology
● Data exchange in compressed format
● Load balance environment

Therefore, it is always advisable to lay the right foundation before you start a build and keep revisiting it as your progress with the product.

Below is the Seattle Conference on Scalability: Scaling Google for Every User