Serverless. Cloud-computing. Agent. All this IT lingo, but what does it really mean? 
It seems like each term has a million different definitions. And I don’t know about you, but when I hear agent I’m thinking, Bond, James Bond.
But I don't think 007 is fighting crime on our computers, so who is this Agent?
What is an Agent?
Let’s do the boring thing first and see what official definitions there are out there. One definition is “a computer program that performs various actions continuously and autonomously on behalf of an individual or an organization”. Another definition specifically for software agents is “pieces of software running on the exhibit device, assisting with, or responsible for, the physical data acquisition. These agents run on the normal operating system of the device and use Application Programming Interface (API) calls for low-level memory access, or they use a dedicated operating system for data acquisition.”
In layman's terms, it’s a lightweight computer program that runs in the background to execute predefined tasks, transparent to the user. I say lightweight because there is no application shortcut (on the taskbar or desktop) for the user to click that will start the program or provide a Graphical User Interface (GUI). It’s always running, starting automatically when the computer boots up. Any potential interaction with the program by the user would likely only occur in the terminal via command line. Use cases for this would be to view logs and other relevant files or potentially stop/start the agent.
Agents can be useful to replace repetitive tasks that occur in the background. They’re also incredibly useful for programs that need to provide feedback to or communication from the internet. Using an agent prevents unnecessary impact on a user with an application that they have no need to interact with. It can run seamlessly in the background to accomplish the tasks required.
Why do we use Agents at Automox?
Without getting into the nitty gritty, there are a couple functions that the agent enables without large impact to the end user. Our agent is a broker of commands sent from our cloud console and will also report back the status. For instance, the console may need to deliver scripts to the device, install software, or force an upgrade/patch to the OS. The agent also provides user notifications to each endpoint. Because these are repeatable, low-level tasks, an agent is a good fit.
Want more resources on the Automox Agent?
Here are a few links on other topics surrounding the Automox agent. Let us know if there are other resources you’d like to see!
Hopefully this was helpful for you! I mean, it’d be cool if Bond was fighting crime on my computer, but we can still pretend, right?