Need Details on ops/sec

Hi,
I am using Search query which gives the result as below.

All,Min Throughput,relative-kibana-discover_50%,0.02,ops/s
All,Median Throughput,relative-kibana-discover_50%,0.02,ops/s
All,Max Throughput,relative-kibana-discover_50%,0.08,ops/s

So, just wanted to understand the meaning of ops/s here.

--Regards,
Balmukund

That is the number of operations of that type run per second. These are reasonably heavy queries scheduled to run at regular intervals, which explains the low number of ops/sec.

Hi Christian,
Thank you for your prompt response. It would be great if you could highlight the operations in the below custom query.

{
"name": "custom_query",
"operation-type": "search",
"index": "elasticlogs_*",
"body": {
"query": {"bool":
{"must": [{"bool": {"should": [
{"match": {"nginx.access.geoip.country_name": ""}},
{"match": {"nginx.access.geoip.country_iso_code": ""}}
]}},
{"bool": {"should": [
{"match": {"nginx.access.user_agent.os_name": "Mac OS X"}},
{"match": {"nginx.access.user_agent.device": "Other"}}
]}},
{"bool": {"should": [
{"match": {"nginx.access.geoip.continent_name": "Asia"}},
{"match": {"nginx.access.response_code": "200"}}
]}},
{"bool": {"should": [{"match": {"nginx.access.geoip.city_name": ""}},
{"match": {"nginx.access.remote_ip": "203.112.82.0"}}
]}}
]}}
}
}
Basically, I want to understand, is it the number of operations of a query per sec?

--Regards,
Balmukund

I am not sure I understand. If you have a look at the track is structured you will see how it is configured.

Hi Christian,
Thank you for your reply.
How do you define operations in the above query? Also, ops/sec means query per sec here?

Please find below operation type.

"schedule": [
{
"operation": "delete-index"
},
{
"operation": {
"operation-type": "create-index",
"settings": {
"index.number_of_replicas": {{ number_of_replicas | default(0) }},
"index.number_of_shards": {{ shard_count | default(2) }}
}
}
},
{
"operation": "index-append-1000",
"warmup-time-period": 0,
"time-period": 1200,
"clients": {{ p_bulk_indexing_clients }}
}
]

Please elaborate here, ops/sec.

--Regards,
Balmukund

The first two operations there runs only once, which means ops/sec does not really mean much. The last one is not rate limited (unlike the queries you mentioned earlier) so that ops/sec will vary. Please have a look at the esrally documentation as this describes the structure of a track and the configuration options.

Hi Christian,
Thank you for your reply. I went through various documents and tried to understand but still have little doubts.I think, below example can make me clear. Sorry, to ask it again and again.

{
"schedule": [
{
"operation": {
"operation-type": "search",
"index": "_all",
"body": {
"query": {
"match_all": {}
}
}
},
"warmup-iterations": 100,
"iterations": 100,
"target-throughput": 10
}
]
}
So, here "search" type operation runs 100 times. It calculates per sec operations from this iterations. Please correct me my understanding.

--Regards,
Balmukund

Here the target throughput is 10 ops/sec which is what esrally will try to achieve. It will not go higher than this.

Hi Christian,
Thank you very much for your explanation.

--Regards,
Balmukund

Hi Christian,
How to check the number of merge threads in elasticsearch?

--Regards,
Balmukund

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