Serverless computing is pegged by many as having the potential to be hugely disruptive in how applications are built and run in the coming years
The name serverless is quite misleading. There are still servers, they will still need to be managed and operated. However, to the end-user developer or IT team they will be invisible – instead they will be integrated and operated within the public cloud instance provided by AWS, Microsoft, Google or whoever you use.
In doing so, this removes a number of the usual considerations for developers such as availability, run time, scalability and server management. In a serverless environment, developers needn’t worry about tuning applications to work with certain types of servers. Everything is done within the public cloud provider.
By removing this part of the process for businesses and offering consumption-based pricing based on event utilisation, meaning the end of paying for idle cloud capacity that isn’t being used, it all translates into massive financial savings for businesses, both in terms of cloud resources and the skills needed to manage them.
Nick Rockwell, CTO at The New York Times believes serverless architecture will be “major shift” that’s a bigger change than moving to the cloud.
Less Cloud, Less Ops
It’s fair to say there is currently a slight lack of enthusiasm for going serverless on both sides. That’s hardly surprising. After all, it will mean many IT operations folks simply aren’t required any more. From the point of view of the cloud providers, they’re taking on a lot more of the work and will be selling less.
“(Cloud providers) make a lot of money selling instances to us that we don’t use. They’re probing and innovating, waiting to see what the reaction of the market is – and I don’t think we’re doing enough to lead them and to validate them on the buying side,” Rockwell adds.
When you add it all up, it’s likely to be a bigger change than many anticipate but getting to that point could take time.