Scripts - Components Of Graphite

So this whole section and all the lectures in this section are about graphite and the stats the and how we install them and how we do different settings on them.

So if you don't want to use graphite you are very welcome to skip the whole section.

Now in this lecture what do I want to talk to you about is the different components of graphite.

So when we install and configure graphite you understand what we are actually doing.

So graphite is made of two major components to reach our graphite web which is the front end and graphite backend or carbon.

It's just called carbon because it goes well with graphite.

So in the graphite web you have a web application and it uses a database to store your graphs all the graphs that you creating graphite where it will be stored in a database.

This database by default these equal light which means that it can be only used as a single instance database because it doesn't support concurrency stuff.

But if you want.

You can use an external database such as in my sequel that is useful when you want to have multiple instances of graphite.

And if you want to create your graphs in graphite web because only graphs are stored in that database for our case because we want to use graph now for creating graphs and we have no intention to create dashboards in graph fight it's 100 percent ok to not use an external database and only use the secret light instance that is shipped with graphite to web.

You can also configure medication D which is a kind of distributed casing component or framework again in case you want to have multiple instances of graphite web.

You may want to use medication to have a distributed case.

I will show you when we finish the installation of graphite.

That how you can use an external database such as my sequel and also how you can enable location in case you have case D already if you use Amazon Web Services you can use a classic case service because it can offer you a best case d and also you can use a W S are the S for a my sequel database.

In case you want to store your dashboards in graphite and you need a database.

This is important when you use containers because your containers may die and then your database will die out with the container so you will need an external database so all the contain there is a store that data outside the container.

But again in our case because we want to use graph on now for the dashboard adding and creating graphs we are not going to use this database.

And on the back inside we have carbon carbon has a component called Carbon case which cases the data that we send to carbon your graphite Web can directly fetch the data from cash.

If you don't really care about the retention of the data and you just want to get it fast but then data should be very recent and you can fetch it directly from the case which is faster or you can store your data in whisper which is a kind of database or data storage for carbon and you can ask data to be fetched from whisper as well.

For example if you say I will send you this telemetry data but keep it for one week then you cannot have that data in the carbon cage for one week.

Therefore it will be stored in the spare database and then therefore data should be fetched from this bear as well.

After we install graphite I will show you that how you can configure everything especially there are two things you need to learn about graphite carbon.

One of them is the retention policy and sample it great and the other one is aggregation.

If you see above carbon capture component there is a an optional component called Carbon aggregator carbon aggregator is very useful when you want to only maintain and hold an aggregation of data not the actual data.

For example if you send millions of data points or telemetry data to carbon but you only care about the average of that data carbon aggregator can aggregate the data into like average based on a given time.

For example average of this data every minute and only a stored that and by doing this you reduce the amount of data you storing whisper and you save disk and you increase the performance.

So let's go to the next section and see how we can install graphite.

So this whole section and all the lectures in this section are about graphite and the stats the and how we install them and how we do different settings on them.

So if you don't want to use graphite you are very welcome to skip the whole section.

Now in this lecture what do I want to talk to you about is the different components of graphite.

So when we install and configure graphite you understand what we are actually doing.

So graphite is made of two major components to reach our graphite web which is the front end and graphite backend or carbon.

It's just called carbon because it goes well with graphite.

So in the graphite web you have a web application and it uses a database to store your graphs all the graphs that you creating graphite where it will be stored in a database.

This database by default these equal light which means that it can be only used as a single instance database because it doesn't support concurrency stuff.

But if you want.

You can use an external database such as in my sequel that is useful when you want to have multiple instances of graphite.

And if you want to create your graphs in graphite web because only graphs are stored in that database for our case because we want to use graph now for creating graphs and we have no intention to create dashboards in graph fight it's 100 percent ok to not use an external database and only use the secret light instance that is shipped with graphite to web.

You can also configure medication D which is a kind of distributed casing component or framework again in case you want to have multiple instances of graphite web.

You may want to use medication to have a distributed case.

I will show you when we finish the installation of graphite.

That how you can use an external database such as my sequel and also how you can enable location in case you have case D already if you use Amazon Web Services you can use a classic case service because it can offer you a best case d and also you can use a W S are the S for a my sequel database.

In case you want to store your dashboards in graphite and you need a database.

This is important when you use containers because your containers may die and then your database will die out with the container so you will need an external database so all the contain there is a store that data outside the container.

But again in our case because we want to use graph on now for the dashboard adding and creating graphs we are not going to use this database.

And on the back inside we have carbon carbon has a component called Carbon case which cases the data that we send to carbon your graphite Web can directly fetch the data from cash.

If you don't really care about the retention of the data and you just want to get it fast but then data should be very recent and you can fetch it directly from the case which is faster or you can store your data in whisper which is a kind of database or data storage for carbon and you can ask data to be fetched from whisper as well.

For example if you say I will send you this telemetry data but keep it for one week then you cannot have that data in the carbon cage for one week.

Therefore it will be stored in the spare database and then therefore data should be fetched from this bear as well.

After we install graphite I will show you that how you can configure everything especially there are two things you need to learn about graphite carbon.

One of them is the retention policy and sample it great and the other one is aggregation.

If you see above carbon capture component there is a an optional component called Carbon aggregator carbon aggregator is very useful when you want to only maintain and hold an aggregation of data not the actual data.

For example if you send millions of data points or telemetry data to carbon but you only care about the average of that data carbon aggregator can aggregate the data into like average based on a given time.

For example average of this data every minute and only a stored that and by doing this you reduce the amount of data you storing whisper and you save disk and you increase the performance.

So let's go to the next section and see how we can install graphite.