Revisiting Server-less
Note: I am rebranding Serverless as Server-less in a brilliant thought leadership move :). You do the same if you want to be seen as cool etc…
I remain a proponent of server-less and believe that it is a big change in how applications will get built. Server-less offers business and technical benefits. Its ideally suited for new workloads. Ignore the vendor lock-in concerns if you want to innovate.
Lets review some of the benefits of server-less:
– Frees up app developers from infrastructure concerns
– App developers focus on business logic, less need for rockstar /ninja/fullstack engineers
— Business pays for exactly what they consume
– Business spend is proportional to business growth
– Get away from instance based pricing, which is a rip off
These benefits of server-less are not possible unless the vendor has done the necessary groundwork. They need massive investments in infrastructure, operations and a well designed programming model.
For now, AWS is the clear leader in server-less. AWS Lambda is still the best game in town and their infrastructure investment is unparalleled. Going with AWS Lambda is a good decision in majority of cases. (One caveat: For some weird reason, AWS has GoPhobia. They do not treat Golang as a first class citizen with AWS Lambda, so you have to use workarounds to get by it. ).
Microsoft Azure Functions is interesting as well and is a good fit if adopting Azure as standard platform. Google is still asleep at the wheels when it comes to server-less. I am hoping they will have some announcements related to Google functions next month.
An emerging project in server-less is Apache Openwhisk. This project offers the last hope for large companies like IBM to survive as a viable cloud company. Its unclear if IBM will do it though. Lately their track record is shameful.
There is also scope for companies to provide services to customers to succeed with server-less. There is need for them as its with any new tech paradigm.
And then we have pretenders. Startups that are trying to recover from betting on a failing stack or a strategy (“private cloud” ). Others where marketing has decided to put the cart before the horse. They hope to stay relevant because server-less is “hot”. They don’t have necessary infrastructure nor operational staff. These companies might talk about hybrid cloud or private cloud or multi cloud and try desperately to tie them to server-less. Ignore them.
We are in year zero of the BS phase of server-less. Mapping what we have seen earlier with other tech, I expect the following to happen in 2017:
– ITIL/Agile/DevOps thought leaders will rush to re-brand themselves as server-less thought leaders
– Some never heard of company will announce a certification program for server-less, AWS will follow suit
– Linux Foundation will be home to a new server-less foundation
– Couple of large companies will announce a Billion dollar investment in server-less
– A VC firm announces a VC fund dedicated to server-less
Be ready for a wild ride of fancy terms and outrageous claims. BS phase of any new technology can wear you down and we are getting there with server-less. Its like “Cloud” in 2010.
That’s all folks. Thoughts?
Source: Revisiting Server-less – Medium