Should Your Enterprise Consider Virtualization Through FaaS?
Function as a Service, or serverless computing, is allowing companies to remove the burden of server management from their IT Teams, freeing them up to innovate.
Function as a Service (FaaS), or serverless computing, refers to a cloud computing code execution model in which the cloud provider manages the starting and stopping of virtual machines to serve requests. The model is also known as “serverless computing,” but this term is something of a misnomer‚ servers are still involved in the execution of requests, but IT professionals utilizing FaaS no longer have to worry about specifying or reserving any server instances.
Amazon, Google, and Microsoft all offer FaaS options: Lambda, Google Functions, and Azure Functions, respectively. These services promise to free up the time of IT professionals with a worry-free method of handling backend functions that allows them to pay only for instances that they actually use, all without any overhead for managing physical servers.
When a developer wants to implement a feature into an application — for instance, generating an image thumbnail when a picture is uploaded, or generating in-app purchase options in a game — they write backend code that is triggered by certain events. Traditionally, managing the infrastructure to run this code required managing, provisioning, and scaling servers, updating and installing security patches, and monitoring the infrastructure’s performance.
FaaS “provides users with unprecedented levels of flexibility, scalability, and operational simplicity,” according to Mesosphere’s co-founder and CTO Tobias Knaup. When using FaaS, the developer uploads the code to one of the aforementioned services, which then manages the capacity, scaling, patching, and administration of the infrastructure. When triggered by an external event, the FaaS runs the snippet of code that corresponds with it; each piece of code carries out a single function.
For example, if a developer writes code for an application that encodes video, he would simply upload the code to a service like Lambda and select the event source to monitor, such as an S3 bucket. When a user uploads a video into the S3 bucket, the event is triggered and Lambda runs the code.
FaaS can be a cost-effective method for managing a company’s server needs. Instead of running applications 24/7 on servers that are always online, an FaaS infrastructure can run requests on-demand. Most companies charge users either per request (generally measured in millions) or per time active (measured by 100 milliseconds). In Lambda’s pricing model, for example, the first one million requests per month are free, charging $0.20 per million requests thereafter.
For example, cyber-security and malware protection company FireEye moved their key analytics functions to Lambda with excellent results. "We saw a massive reduction of costs associated with the analytics," FireEye’s operations engineer Joseph Kordish explained to TechTarget.
Perhaps the greatest benefit of moving to an FaaS infrastructure is the ability to free IT professionals from the worries of server management. In the words of Senior Solutions Architect for CorpInfo Laith Al-Saadoon, "serverless computing… allowed the team to focus on functionality and features over provisioning virtual machines and operating systems.” Eliminating the tedium of "undifferentiated infrastructure burdens… can really free your teams and, therefore, the business to innovate and iterate at an unmatched rate, providing an edge on the competition," he said.
IT professionals are a crucial asset to any company, and for that reason, FaaS is a solution that companies should certainly consider implementing. The strategy might allow these teams to pursue more productive tasks, like developing new applications and improving functionality across the board.
But it isn’t just FaaS that can free the time and resources of IT professionals: with the right capacity planning tools at your disposal, you can ensure that your IT department has everything it needs to do what it does best. TeamQuest’s Vityl Adviser helps IT teams anticipate when, where, and how IT resources will be needed to meet future demand. With this powerful information at your disposal, you can avoid costly downtime and save hundreds of work-hours.