Scaling based on average load

October 9, 2013 2k views
I want to scale my application (by opening a new droplet) based on the average load reading. Is it a good idea to use this approach? What will be a high load value for a 1 core droplet? Above 1? But for a 2 cores droplet will be above 2?
4 Answers
Unless you're getting 1,000 hits a day, you don't need an upgrade.
by Bulat Khamitov
With DigitalOcean’s simple interface with API access, and full root level access, here is a great opportunity to create an automated setup that scales itself on-demand. You can have your websites scale up or down automatically to meet the traffic demands.
I have read the articles, but they are more focused on 'how to scale' then 'when to scale'. I'm planning to move to DigitaOcean a php application that needs more processing power than a regular website. It is running on a dedicated server now but it becomes overloaded on traffic spikes. I need a scalable solution. I want to make the php app to be self-aware when the host is overloaded. I'm looking for a simple solution and I was thinking to use the server load averages but I'm not sure what value would indicate that the droplet is overloaded and based on this, to open a new droplet. My questions again:

Based on the load averages, above which value should I open a new droplet?
Do you think I need more advanced monitoring tools for this or the server load average reading would be enough?
It depends on what resources your application uses the most, e.g. RAM, CPU, Disk IOPS, etc.

Unfortunately I don't have experience with auto scaling however I'm sure you can create a bash script that reads the load averages and uses the bash script from that article to scale up if needed.
Have another answer? Share your knowledge.