Are there any limitations that the Elasticsearch can monitor only these many servers? My understanding is, there are no limitations and it really depends on the Nodes configuration. Please let me know in case of any changes.
In addition, i tried to understand on how to size my servers' to monitor 50 servers'. I will be monitoring the system logs and application logs and the server metrics. I would like to have the data retained for 2 months time.
Is there a "Sizing calculator" that will help me size my servers' for above requirement? or could someone pls help on how the configuration should be? Like how many Nodes, CPU, RAM, Storage etc.
@ganesh2, hardware is generally based on how much logs are at your end. If you have 50 Servers and generating total 10 GB data per day at your end. Then it's size in elasticsearch will be around 10 times (100 GB) if you use one replica.
Replica is copy of your data. If your actual data size in elasticsaerch is 50 GB then 50 GB will be replica so its become 100 GB. The size of data in elasticsaerch is increased because it store the data in structured form and add multiple fields to data.
So from above you can calculate how much disk size you will require.
In starting phase as per my knowledge you can start with 3 Node. You can use all three node as master as well as data node, but its not good practice. You can use 8GB RAM, 4 vCPU servers.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.