- What trade-offs are you making when using the cloud? Have you consciously evaluated these trade-offs? What is your justification?
When using the cloud, you are giving up some control of your data. For instance, you likely do not know the physical location of your data–it is abstracted so much that you may know the “zone” it resides in throughout the world but not much else. When using your own server, you likely have physical access to the hard drive and can ensure its physical security is under your control. Cloud companies make claims about how secure your data is and that there is little downtime–but there are still certain risks associated with that. For instance, in managing my own server, I can be sure I am in complete control of uptime and downtime and maintenance, whereas I may not be able to do anything when an outage occurs on other services.
I have consciously evaluated these trade-offs and think using cloud services makes so much more sense than personally setting up and managing my compute environments. I am afforded so much more power at a lower cost and have the ability to expand easily using load-balancers. I know GCP and AWS hires engineers full time to keep my data secure and running 99.999% of the time.
- Is it ever worth it to manage your own private cloud services? Do you envision a future where you may use your own services rather than third party ones?
It is not worth it for me to manage my own private cloud services. I build lots of websites for people using WordPress and rely on services large companies can offer that I just cannot do myself. For instance, GCP and AWS have vendors who offer one click set-up for WordPress sites, DDoS protection, and automatic scaling as the websites need to scale up or down.
I cannot see myself using my own services rather than third party ones in the future. I think the use case has been well-tested with many Fortune 500 companies and the cost benefits and security insurance outweigh the benefits of setting up and using my own services. Consider that for this project, we set up GitLab via GCP…imagine if we had physically built our own server and installed necessary components from the ground up! The point is, there is convenience in abstraction and I simply do not have the time, will, nor reason to be constantly reinventing the wheel.
- Do you have the moral standing to complain about encrouchment on your privacy when you consciously give away your information to third party services?
These services operate through contracts, so Google or Amazon legally obligate themselves to certain standards. I believe when those standards are breached, I am justified in complaining. It’s important for these services, however, to display transparency in communicating with consumers so that each side understands the expectations. If I consciously give my information (and money) to Amazon or Google, I also have to expect privacy and security from them in return.