Learn about the value of short terms cloud analytics.
- [Instructor] So, let's talk about short-term monitoring analytics. This is reporting, so we need to monitor systems on an ongoing basis and provide reports as to what occurred, either short-term, as well as long-term. System logs may be part of compliance. It may be a legal issue to understand that we need to monitor everything that occurs within a particular system over a period of time and be able to report those to some government agency, as the law dictates. We need to correct issues as we find them.
So ultimately, this is not just about creating reports, but this is about leveraging that information to correct things as they pop up. So, back to our framework. Again, we have reporting between governance and health. Core to that is the capability of laying down, to some sort of persistent storage, a log, as to what's occurring within the particular system. A niceidy, or something that's added onto this, is the ability to take that log, load it into some sort of a database and understand the trends that are occurring over time.
Is performance going down? Is service levels going down? Is storage being depleted quicker than we thought? Those sorts of things need to be understood and only long-term and short-term reporting is going to provide you with the capability of doing that. So, short term reporting is what's running again. We need to figure out what's occurring, applications, databases, all sorts of things that are in the mix and what those things are doing and what key metrics do we need to lay down in the log.
How well are they running? In other words, are they meeting expectations? Are they leveraging the resources effectively, in order to support the particular functions that they carry out? Any corrective actions need to be taken if they fall out of a particular parameter. Again, this is short-term reporting. So, this may be a couple of weeks, a day, maybe even last hour. We need to understand that this information is checked on shorter intervals and corrective action taken, if it needs to take place.
So again, back to the technology needed. We need data. This is new, we need to be able to stick information as to what's occurring within the system within some sort of a database. Either that's going to be on a flatfile log, such as a text file, which is the traditional way of doing performance monitoring and monitoring reporting, but these days, we have a tendency to stick them into object-based databases, relational-based databases. So we could do more with them. We can trend them, we can mine them, we can understand the data in the context of other data.
We can slice and dice it, to understand better what the systems are doing short-term and long-term. We need to understand the trends. Understand the limitations. In other words, what are the limitations that we're setting, basically policies that we're constructing, in terms of how the particular Cloud providers and the applications and the resources that are running on the particular Cloud providers may go out of some sort of limitation we set and we have to do something to correct that. Actions that are taken. In other words, if we're out of limitations, what actions do we need to take to correct things? Then reporting.
The ability to generate reports out of these systems as need, to do whatever is needed, to make sure that we find the issues and correct the issues before they become issues.
- Cloud health, performance, security, and governance monitoring
- Cloud monitoring analytics
- Cloud monitoring costs
- AWS CloudWatch
- Librato CloudWatch
- Cloud Cruiser
- Microsoft cloud monitoring
- Rackspace cloud monitoring
- Creating a cloud monitoring and operations plan
- Defining cloud monitoring operations patterns