“One of the biggest benefits of serverless is that it allows for rapid development and deployment, which results in faster time-to-market. Serverless computing essentially enables developers to significantly reduce the development cycle, thereby reducing costs for an organization,” says, Anshu Agarwal, CEO and co-founder, Nimbella
ET Bureau: What are the different areas that serverless computing is being used in the current situation?
Anshu Agarwal: Serverless computing encompasses a broad range of offerings. At the core, technology allows developers to build applications more efficiently and effectively without needing to worry about infrastructure management.
Today, there are many use cases of serverless computing. Some of the most common ways it is being used include Websites and APIs, Event Stream Processing at Scale, Multimedia Processing, and IoT Events.
ET Bureau: What are the main motivating factors for technology executives to invest in serverless computing?
Anshu Agarwal: Leveraging serverless computing is critical for teams that are looking for an efficient way to scale quickly and reduce operational costs. Many technology executives turn to serverless as a strategy to help them innovate, build better products to address market demands, and ultimately stay competitive.
One of the biggest benefits of serverless is that it allows for rapid development and deployment, which results in faster time-to-market. Serverless computing essentially enables developers to significantly reduce the development cycle, thereby reducing costs for an organization.
It also takes away the need for server management, reducing the development team’s time to manage infrastructure. Once we consider all the human capital efficiencies of serverless computing, the cost savings can be very significant. After all, the cost of developers’ time is often the highest cost to a business.
There are also operational reasons why technology executives want to invest in serverless computing. For example, a growth-stage company that’s scaling quickly will need agile, adaptable technology and doesn’t become a roadblock for its changing needs.
Serverless architecture is inherently scalable, which allows companies to serve their growing number of users efficiently. A serverless application can handle an unusually high number of requests just and process a single request from a single user.
On the other hand, a monolithic application with a fixed amount of server space can crash by a sudden increase in usage. Furthermore, serverless applications only use resources when the application needs backend functions, and it automatically scales up as needed. Provisioning is dynamic and real-time, thus reducing operational costs.
ET Bureau: How are Function-as-a-Service and serverless computing connected?
Anshu Agarwal: Function-as-a-Service (FaaS) is a subset of serverless computing, although they are used interchangeably. In my opinion, FaaS was the first milestone in serverless because it offered to compute.
The serverless computing paradigm has evolved to include serverless memory and storage and function-to-function networking because most applications don’t just require computing. They require three pillars: compute, storage, and network.
A function by itself isn’t an application, and real-world applications mostly have state, and the functions need to communicate with each other. There are many forms of state. If you are building an API, and it has a front end component and a web component, you need storage, and the kinds of storage you need are different.
If it is an application with a front-end UI, you upload those assets to an object store, as one storage form. When handling session state or transient data streaming properties, you might use a key-value store. If you are managing files, and your application needs to process images or PDFs, etc., you need a file system.
These are needed for developers to build a stateful serverless application. Therefore, for serverless computing to be applicable for various workloads and applications, it has to have more than FaaS.
ET Bureau: How can tech executives leverage serverless computing to empower their developers to build more scalable cloud-native applications?
Anshu Agarwal: Tech executives can leverage serverless computing by adopting a serverless architecture and a serverless platform for their applications. This allows their developers to only code business or application logic, leaving the platform provider’s infrastructure burdens, whether it’s in the public cloud, private cloud, or hybrid.
But this requires adopting a serverless platform that serves the need for stateful applications because most enterprise applications have a state. The reason why a state should be supported natively is that it is necessary to maintain locality.
In serverless and containers-as-a-service, there is no locality. One doesn’t know where their functions are running or where your containers are running, but you need access to data, so if you can co-locate compute and data close together, you have sub-millisecond latency. You can take advantage of system-level caching to help reduce your costs.
And that translates to is better application performance. Getting back to what serverless promises is really about freeing the developer from thinking about various things. If you want a key-value store, you should be able to use it without thinking about where it’s provisioned, how it’s provisioned, and how it’s backed up.
There are basic technologies that are sort of just legacy tech that the platform must provide. And then there are features that you can’t do at the user level. You can’t optimize your cache, and you compute because proprietary platforms don’t give access to that.
Read More: Negotiating a Job-Offer in Uncertain markets
These are the things that the underlying platform has to do, and that’s what tech executives should look at while considering a serverless platform. The serverless architecture is wonderful because it ensures the inherent scalability and native state management directly translates into better application performance.
Anshu Agarwal is the CEO and co-founder of Nimbella, a serverless cloud provider. She has over 20 years of experience in the technology industry, building and evangelizing products that have solved real-world problems for organizations large and small across the globe. She served on the Board of Directors of the OpenDaylight Project, Linux. Prior to Nimbella, she successfully led product and marketing teams at four infrastructure startups: Cedexis (acquired by Citrix, 2018), ConteXtream (acquired by Hewlett Packard Enterprise, 2015), Ankeena Networks (acquired by Juniper Networks, 2010), and Speedera Networks (acquired by Akamai, 2005). Anshu holds a B.S. E.E. from IIT/Roorkee, India; M.S. E.E. from Rutgers; and an MBA from the Kellogg Graduate School of Management.