Comprehensive infrastructure monitoring is a basic requisite for all IT systems as when issues arise it is necessary to have detailed information available as to what is occurring within the technology stack in order to quickly pinpoint the root cause of incidents and to restore services in the quickest possible timeframe.
This is achieved through monitoring the infrastructure elements that contribute to the delivery of a business service or application. Almost all IT departments have some level of infrastructure monitoring in place though to varying degrees of effectiveness.
Fundamentally, there are two primary technical methods for obtaining performance and availability metrics from infrastructure:
- Agent Based
Involves the installation of software (“agents”) on the infrastructure being monitored and enabling these metrics to be sent back to a central repository for actioning. Monitoring solutions of this type will generally be able to provide a richer and deeper set of monitored information and is not dependent on the delays associated with ‘polling’ technologies, however does come with the added overhead of maintaining the distribution and versioning of this software. - Agentless
This solution is titled ‘agentless’ as is does not involve the deployment of software onto monitored infrastructure, but instead knows how to ‘connect’ to that infrastructure and polls it at regular intervals to obtain performance and availability information. Although significantly reducing the administrative overhead of managing the solution as reduced software is deployed, in general the depth and richness of the information obtained is less than through agent based systems.
Our experience has been that there is no single solution that is best for all customers – and a number of factors contribute to determining the most appropriate solution. In large complex environments the required solution will often comprise a combination of both agent and agentless technologies, usually from a variety of vendors specialising in particular technology platforms.
JDS has a lot of experience with infrastructure monitoring and can help with advice on what method works best for an individual situation, what level of monitoring works best and how to bring the information together to provide a coherent understanding of how the infrastructure relates to the business services being delivered.