This is the final post in my blog series addressing the many myths and misconceptions about agent-based IT asset management technology. The four parts to the series are as follows:
Part 4: Network and End-User Impact
Another common fallacy about agent-based technologies is that they create a significant burden on the network. In the past, available bandwidth was, in fact, a legitimate concern; but network capacity has evolved at a rapid pace and most IT asset management vendors produce an agent that’s small and streamlined—and in some cases even configurable in terms of bandwidth consumption. In reality, software inventory agents represent a negligible load when compared, for example, to the amount of data transmitted by a user surfing the internet. And while we’d expect there to be an initial network impact when deploying the agent for the first time, once the agent has been deployed, the network impact should be no different from an agent-less solution collecting the same type and volume of data. After all, why would it be?
Related to this, some administrators express concerns that installing agents on end-user machines will compromise system performance or utilize too much memory. This is also an historical misconception based on an assumption that computers still have miniscule hard drives and little processing power. In addition, people are often resistant to installing “another agent” because they assume that asset management agents resemble much more invasive agents, such as those belonging to antivirus, spy ware, and imaging technology.
But the truth is that most IT asset management agents are very small Windows applications that perform the same operations on the local machine as would be invoked if issued remotely. Such agents are designed to run at low priority to minimize system impact and, assuming they are provided by a reputable vendor, will run silently and invisibly on end-user machines.
So how does one determine the right approach for their organization? First of all, before accepting the notion of “invasiveness”, it’s important to ask exactly what that means, and what, if any, the true day-to-day implications are. Aside from the initial time and overhead associated with deployment and upgrades, I’d suggest there are very few, if any, negative impacts related to the presence of a software inventory client , assuming the client is delivered by a competent vendor.
Second, it’s critical that anyone researching asset management technology thoroughly acquaint themselves with the current generation of agents. Agents are not the bloated, clunky workhorses of the 1990s that belonged to large framework solutions; today’s agents are small, nimble, and occupy a negligible footprint.
Most importantly, don’t lose sight of your business and IT goals as you evaluate the different ITAM solutions—the deployment methodology shouldn’t drive your decision. Agent-less deployment is simply a feature—not a means to an end.