New droplet or droplet upgrade?

February 27, 2018 161 views
LAMP Stack Ubuntu

I am not yet in this situation, but it won't be long...

I am running an Ubuntu LAMP server with a few instances of wordpress on it. As I keep getting new clients, obviously I am going to eventually need to either upgrade the current server or spin up a new one to keep the user experience acceptable. Either my Google-fu is weak today or there isn't anyone asking this question, so I thought I'd pose it to my peers here.

Basically what I am wondering is this - is there a compelling reason to go with either a server upgrade or new server droplet - the dollars work out pretty much the same no matter what I do, so I'd like to look at it from a standpoint of performance benefit - or maybe ease of maintenance... or just whatever!

Anyone out there made this decision before?

2 Answers

This really is a question of preference really and what the needs of your client(s) are. With running a single droplet and all the sites in one regardless if you resize it up you are running into a ceiling in regards to performance at some point.

You may or may not see it, however if you do, say one of the sites all of a sudden begins to get tons of traffic and consumes allot of the single droplets resources, the other site(s) may also see a decline in performance.

Using two droplets would spread the load between both and only have some sites effected in the above scenario.

Another solution is to perhaps look into a distributed setup making use of something like Docker

by Mitchell Anicas
When deciding which server architecture to use for your environment, there are many factors to consider, such as performance, scalability, availability, reliability, cost, and ease of management. Here is a list of commonly used server setups, with a short description of each, including pros and cons. Keep in mind that all of the concepts covered here can be used in various combinations with one another, and there is no single, correct configuration.

Thank you for the response, sir... I had considered that distributed setup as well, but hadn't yet done the research into the work I'd have to put into it. It sounds like it might be a decision best saved for when I see what's actually causing the slowdown (i.e. bandwidth, RAM, or something else) - it's looking like right now the LIMFAC is going to be memory - I have CPU and bandwidth to spare, but as you pointed out that could change at any time. Thank you for the link - I'm going to add that sort of solution as a possible as well.

Have another answer? Share your knowledge.