One of the huge problems with a 'big box' solution is the lack of statistical an...
One of the huge problems with a 'big box' solution is the lack of statistical analysis. Alarms are a passive way to manage a network - you wait for an alarm to occur and then react to it as needed. It could be said it's only 50% of a solution - the other 50% being statistical monitoring or active monitoring. When polling statistics, you can trend for when you need new hard drive space, or hardware in general, the limits on your current capacity or try to alarm when problems could start developing before they actually do.
For my current employer, doing statistical management actually consumes more of my time than anything else - it's also very important to management. I'm using a few open source packages to get the job done since there doesn't seem to be an acceptable solution readily available.
Has there been any thought about adding something for this in Rivermuse?
RiverMuse made a conscious decision to initially focus on event management altho...
RiverMuse made a conscious decision to initially focus on event management although we do believe that the platform is suitable for a much wider application. As a new company you need to be careful not to spread resources too thinly and fail to make progress in any direction. There is an English expression, "Jack of all trades and master of none", which sums up what I am trying to say.
There has been interest from a community member who is considering building a PM application on top of the RiverMuse Core product and we wait with interest to hear more about this.
We do have a sister company called Prelert (http://www.prelert.com/) which may be of interest to you.
The RiverMuse product is of course agnostic in terms of what it manages, so if your PM system is capable of generating performance based events you can of course manage them within RiverMuse.
Comments (2)
04 Nov 2009
Jacob Steinberger says:
One of the huge problems with a 'big box' solution is the lack of statistical an...One of the huge problems with a 'big box' solution is the lack of statistical analysis. Alarms are a passive way to manage a network - you wait for an alarm to occur and then react to it as needed. It could be said it's only 50% of a solution - the other 50% being statistical monitoring or active monitoring. When polling statistics, you can trend for when you need new hard drive space, or hardware in general, the limits on your current capacity or try to alarm when problems could start developing before they actually do.
For my current employer, doing statistical management actually consumes more of my time than anything else - it's also very important to management. I'm using a few open source packages to get the job done since there doesn't seem to be an acceptable solution readily available.
Has there been any thought about adding something for this in Rivermuse?
05 Nov 2009
Ian Best says:
RiverMuse made a conscious decision to initially focus on event management altho...RiverMuse made a conscious decision to initially focus on event management although we do believe that the platform is suitable for a much wider application. As a new company you need to be careful not to spread resources too thinly and fail to make progress in any direction. There is an English expression, "Jack of all trades and master of none", which sums up what I am trying to say.
There has been interest from a community member who is considering building a PM application on top of the RiverMuse Core product and we wait with interest to hear more about this.
We do have a sister company called Prelert (http://www.prelert.com/) which may be of interest to you.
The RiverMuse product is of course agnostic in terms of what it manages, so if your PM system is capable of generating performance based events you can of course manage them within RiverMuse.