ES Benchmark for 2.4.5 and 1.5

Hi All

I want to take elastic search benchmark for 1.5 and 2.4.5 cluster,i am tring to use esrally,i dont know whether this esrally tool will support or not,
can you please suggest me which tool i have to use? do u have any sample please share here

if any easy steps to take benchmark mean please shareme


Hi @Thennarasu_Subramani,

Rally should work fine in your case. Please checkout the documentation at which helps you to install Rally and get started.


Thanks @danielmitterdorfer , i am new to this what are the steps i have proceed ,can you elabrate here?

some of the forums are says please use elastic search 5.0 and above ,thats why i am little bit confussing the elastic 1.5 and 2.4.5 support or not

Hi @Thennarasu_Subramani,

it depends on what you want to do but I can just repeat what I've said before: ES 1.5 and 2.4.5 should work fine.

The Rally docs are very elaborate. Please check there first and I am happy to help out with specific questions. If you are in a hurry then start with


@danielmitterdorfer it taking much time for configuring ,one day gone .this is for one time setup or every time taking benchmark it take this much time?

e for this setup,

Hi @Thennarasu_Subramani,

I am always happy about concrete ideas and suggestions about improving the getting started experience. Can you explain what was so time-consuming? After the installation, Rally auto-detects / auto-configures as much as it can so the basic configuration should be done within a minute.

Usually configuration needs to be done just once; after that you just run benchmarks. As you're running on EC2, obviously you need to do the setup routine every time you create a new instance but you can (and should) automate that part anyway.


Hi @danielmitterdorfer

The above bench mark its taking geoname based, i dont want this matrix

I am having one Es server one data node and master node and no replica ,here i am just indexed 10 million data i want see this benchmark matrix .can you please give me idea how i have to do this

the above one its taking defalut data and providing matrix

Hi @Thennarasu_Subramani,

I am not familiar with the term "matrix" in the context here but I think you are calling a benchmarking scenario a "matrix" (never heard of that term...).

Based on that I assumption you want to create something that is called a "track" in Rally (i.e. a benchmarking scenario). In the docs you will find:


Sorry @danielmitterdorfer i am not understand for track creation ,there having one of json file ,in my case i am having 10 million data in server ,there i need to take benchmark,i dont know how to configure my own track,can you please do some more example

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.