Jizaboz wrote: Thu Aug 25, 2022 9:44 am
Mainly just because I spent so many years (10+ now) working with AWS. I'm a lot more used to the tools and find the documentation to be a lot more relevant for what I need to want to do as opposed to Azure. There are those who say "why not both?" but I think it's better to just stay efficient in the platform you are comfortable with.
Customer: What's this new thing I'm reading about, Docker and cubr-er-neats?
Consultant: It's pronouced 'ku-be-net-eze. Say it wrong and salesmen will know you're a rube and can overcharge you. Naah, it's not important, it's just another fad, there's nothing wrong with chroot jails and other well-known isolation techniques. I've been using them for over a decade and see nothing worth changing to.
Listen, if I can offer you advice, if you want to stay marketable, keep learning. Get out of your comfort zone. See what other offerings are out there. Maybe you\ll have a customer who wants high-availaility and -redundancy. The answer might not be to spin up Kubernetes / Docker instances on AWS East
and AWS West, maybe you set up on AWS East and Google Cloud. Or AWS West and Microsoft Azure. Or maybe Amzaon's offerings are not a good fit and the client should use some combination of Google, IBM, and Microsoft. Or someone else that is offering something better than AWS or ant of the others.
If you know more than one, you can tell a client when splitting services is more cost-effective or provides better redundancy. Or, if they can be split on load factors, the customer is on a lower service tier on each one and it costs less. Or that, given the circumstances, one of them is a much better fit for the customer.
There is the story of one client that wanted to find out why their AWS bill was so high. He discovered that they were using quite a lot of server resources. So, buy 50 used 16-core machines for less than $30,000, use them for compute-heavy applications, and get Internet from two different providers. Then have fallback-on=failure to AWS and use fewer instances with lower resources. Company's $40,000 a month AWS bill drops by half with no change in workload. After the second month, the savings paid for the changes. There is a slight increase in electricity use, but not enough to be concerned. This was the most effective way to make a big change. Other changes to increase savings had tradeoffs he was sure the customer would not be willing to do at the start, but could be persuaded once they saw the results.
Knowing your customer and knowing what choices there are gives you more options and alternatives to offer the customer. If you just keep doing the same thing, yeah, sure, you can become expert in it, but if what they offer is not a good fit and the customer finds out you weren't aware of it? it is likely to make them not be your customer anymore. Then, if they hear about something else, you can tell them you knew about it, but didn't mention it, because while yes, it's cheaper, they have these issues: X, Y, and Z. Or you’re proactive, you tell them about it and explain why it's not a good choice for them. Or that it is.
The better educated you are, the more ways you can find to serve a customer's wants and needs. Which means you make more money.
