This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

Monitoring Services as opposed to components

Hi. Looking for some good reviews on this product. As an example, my SLA  as ICT with other departments is service based and not component based. By this I mean, other departments do not care what comprises network. All they want is x.x% network uptime over period Y. At a technical level, I need to monitor all components that contribute to network up-time-switches, routers etc and proactive take action should service levels start going down, before a dowtime. The other example I can give is say in a banking environment, if I look at ATM or Internet Banking or Mobile banking as a service, there is a whole lot of components that make up such services-Databases, Applications, network  and how the customer is served in terms of were there request timeouts basically customer experience.

1. Is this a product that can address the above ie. monitor services and not components with reports being presented in dashboards?

2. How easy or difficuly is it to implement the above?

3. Anyone here with experiences to share around this?

Thank you!

  • What you could do is a create a group and add all the components or applications that will make up this service.

    Create groups

    Here is an example of a group monitoring IIS and Appinsight for SQL

    pastedImage_1.png

    pastedImage_2.png

  • From experience your issue is going to be getting someone to tell you in advance what constitutes a "down" or "degraded" service, or more importantly what doesnt constitute that.

    If you've got a clean set of rules you can program that into groups. You are unlikely to get a clean set of rules. This is a human problem not a SLW one

    On the SLW end it's difficult to adjust availability numbers in post if there's a disagreement of any sort.

    I'd say overall it's fiddly, depends though. If you've got simple services and simple rules the results will be simple to track and work with. If they're complicated and you've got proper resources and information that's OK too.