A to Z of Google Cloud Platform a personal selection — z — Zero Ops
This is the last entry in my series and I had a number of topics I could have chosen to write about only I ended up talking about them elsewhere in the series or I felt I couldn’t add anything to the topic so I’m going out opinionated.
Firstly I’ll just state that I do not believe how much of a managed platform you have there is no such thing as Zero Ops as you will always be looking at logs (these include audit logs!) , responding to issues, managing security controls etc . Using the Cloud does not absolve you of that sort of responsibility so someone in your org needs to do that whatever label you deicide to use for those activities!
Okay now that is out the way what do I mean when I use the term “Zero Ops” ( as if I ever would use it anyway , it’s in the same bucket I keep the term DevOps these days ) :
Not having to worry about cuddling tin ;not investing in tasks that provide no benefit for your business focus; not worrying about compute sizes, sharding strategies; disk perf etc because all that stuff is taken care of .
[ An aside ] I believe the latest buzz term in the industry I work in has moved on somewhat from No Ops/ Zero Ops to Serverless and that seems to be irritating folks as much as the abuse of the term DevOps and No Ops has been !
GCP has plenty of services that fall into the Zero Ops box. It’s nothing new to them it just is . They may have been a bit ahead of the curve I feel.
So here’s my personal list of GCP services that fit my definition of Zero Ops today with accompanying Zero Ops strapline :
BigQuery — Fully managed, petabyte scale, low cost analytics data warehouse. BigQuery is serverless, there is no infrastructure for you to manage and you don’t need a database administrator to tweak performance parameters.
Dataflow — Transparently handles resource lifetime and can dynamically provision resources to minimise latency while maintaining high utilisation efficiency. Dataflow resources are allocated on-demand providing you with nearly limitless resource.
Datastore — Automatically handles sharding and replication, providing you with a highly available and durable database that scales automatically to handle your applications’ load.
App Engine — Scales your application automatically in response to the amount of traffic it receives so you only pay for the resources you use. Just upload your code and Google will manage your app’s availability. There are no servers for you to provision or maintain.
Datalab — Built on Jupyter (formerly IPython) , runs on Google App Engine and orchestrates multiple services automatically so you can focus on exploring your data.
Pub/Sub — Fully-managed real-time messaging service that allows you to send and receive messages between independent applications
GCS — Durable and highly available object storage
Cloud Functions — Lightweight, event-based, asynchronous compute solution that allows you to create small, single-purpose functions that respond to cloud events without the need to manage a server or a runtime environment
With those services alone you can build some pretty funky solutions without ever having to worry about the nuts & bolts of lower level infra services . Someone else worries about keeping those services up and running. If you’re interested in how that is managed as you know someone has to do that stuff just not you as the consumer of the “Zero ops” (sorry !) service then have a read of the SRE book .
So I’ve done it I managed to have an entry for every letter of the alphabet despite Y challenging me somewhat. I posted one a week for 24 weeks ( E was put under A as it was a logical place for that entry and I posted 2 in one week early on hence why 24 and not 26 weeks ). I’ve enjoyed writing the series and I hope that you have found something useful, or at least some food for thought amongst the selection.
For a little bit anyway I’m dropping the digital pen here for a bit ( Yeah I know I did end the series like that !) .