Many software vendors require agent installs on their customer’s servers as a part of their solution installs. Some need them to collect metrics, while others need them for security, management and other reasons. Agents typically run as a service on servers, and they consume resources, CPU and memory.
Mid-sized and larger organizations can have several different vendor solutions that require an agent install. Existing approaches to agent development and focus have a number of issues. The first is that the majority – if not all – of vendor agents are single use, meaning that they only function for their respective solution. The problem is that if an organization has five solutions that require an agent install, they could potentially have five different agent installs on their servers. Each of these agent installs are essentially providing the same general function, and each are consuming server resources providing them. It is duplicative, a waste of resources, and a drain on IT.
The second issue, which is far more critical, is quality. The development of agents can be somewhat of an art, and is generally not the primary focus of most software vendors. Most do not place a high emphasis on them, and as such they do not develop high quality agents. The result is that many agents are poorly designed, can be resource hogs, and in some cases be unstable. The instability can result in server crashes, degraded server performance, both of which can directly affect application availability. If you have a couple of poorly developed agents that are adversely impacting your critical applications, you have a big problem.
The solution is that software vendors need to get out of the business of developing agents. The majority do not do it well, and their poor quality negatively impacts IT operations, application performance and availability. The acceptance of single purpose agents needs to stop. They needlessly waste resources and impact application availability. Organizations need a single multi-purpose agent that can support multiple functions and applications. An agent that is efficient, stable, reliable, and can be customized by them or software vendors to provide the services they require. A single multi-purpose agent will consume less resources, and will be far more stable and reliable for the customer and their applications. The vendors benefit by being able to focus all of their resources on their core solution.
Those of you that are familiar with Vallum Software know that our solution utilizes an installed agent for deeper metrics, and the installation of Halo Apps. You may also know that our agent, the Halo Agent, is multi-purpose, programmable, and open sourced. What you may not know is that while it functions with the Halo Manager solution, it can also function with other vendor solutions. Vallum developed a software development kit (SDK) that customers and vendors can use to customize the Halo Agent for their needs. It is a win-win for both customers and software vendors. If you currently have the Halo Agent installed on your servers, talk to your other vendors that utilize agents and encourage them to support and use the Halo Agent. It will save you server resources, and possibly avert application slowdown or downtime.
About the Author:
Lance Edelman is a technology professional with 25+ years of experience in enterprise software, security, document management and network management. He is co-founder and CEO at Vallum Software and currently lives in Atlanta, GA.