Back in December 2005, a small company, led by former PayPal employees was launched. This is a story of YouTube.
During the first month, YouTube was serving 8 Million video views per day. This required only a small amount of servers to accommodate and stream the media. When traffic increased and content size exploded, the Ops team had drastically increased the number of servers it possessed in order to deal with the growing volume of requests.
Today, every Cloud Architect fabricating an application architecture know the solution to such catastrophe. Dealing with the production workload does not require servers anymore. Designing applications in a serverless architecture provide you with that exact solution for scaling while minimizing the level of attention required for operating a production environment.
Bustle.com, for example, operates two separate media websites with more than 52 million visits a month and reportedly handles over 100 million events per day. At this traffic size, the 12 APIs Bustle maintains are constantly under stress and must be monitored for any malfunction or hiccup.
That’s where Tyler Love brought Amazon’s serverless platform—Lambda—to the table. Encouraged by their experience with other AWS Services—Kinesis and API Gateway—Bustle could easily migrate its complete production environments to serverless using Lambda. DevOps teams accustomed to managing and adding new servers were immediately liberated from these duties.
Tyler Love opted for AWS Lambda since it was more suitable for considering the requirements of Bustle. But how do you choose which serverless platform is suitable for your business?
Also, the chances of choosing a wrong serverless provider are high!
While choosing the serverless provider, there are popular opinions which are based on various biases, however, there are things that go beyond the surface level.
- Language Support & Deployment
- Dependencies Management
- Persistent Storage
- Identity & Access Management
- Triggers & Type
- Concurrency & Execution Time
- Scalability & Availability
- Logging & Monitoring
- Pricing Model
- Platform Performance
Here's something which can help you: AWS Lambda vs Microsoft Azure vs Google Cloud Functions: Comparing Major Serverless Providers. Let's consider their technical capabilities which can help you in equally weighing the selection process required for your specific needs.
Feel free to share your experience and suggestions. Why did you choose the particular serverless provider? Connect with on Twitter @jsitapara to talk anything about serverless, AWS Lambda and anything in-between.