lucene-solr-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Andy <angelf...@yahoo.com>
Subject Re: Benchmark Solr vs Elastic Search vs Sensei
Date Fri, 27 Apr 2012 20:55:37 GMT
What is the performance of Elasticsearch and SenseiDB in your benchmark?


________________________________
 From: Volodymyr Zhabiuk <vzhabiuk@gmail.com>
To: solr-user@lucene.apache.org 
Sent: Thursday, April 26, 2012 9:50 PM
Subject: Benchmark Solr vs Elastic Search vs Sensei
 
Hi Solr users

I've implemented the project to compare the performance between
Solr, Elastic Search and SenseiDB
https://github.com/vzhabiuk/search-perf
the Solr version 3.5.0 was used. I've used the default configuration,
just enabled json updates and used the following schema
https://github.com/vzhabiuk/search-perf/blob/master/configs/solr/schema.xml.
2.5 mln documents were put into the index, after
that I've launched the indexing process to add anotherr 500k docs. I
was issuing commits after each 500 doc batch . At the
same time I've launched the concurrent client, that sent the
following type of queries
((tags:moon-roof%20or%20tags:electric%20or%20tags:highend%20or%20tags:hybrid)%20AND%20(!tags:family%20AND%20!tags:chick%20magnet%20AND%20!tags:soccer%20mom))%20
OR%20((color:red%20or%20color:green%20or%20color:white%20or%20color:yellow)%20AND%20(!color:gold%20AND%20!color:silver%20AND%20!color:black))%20
OR%20mileage:[15001%20TO%2017500]%20OR%20mileage:[17501%20TO%20*]%20
OR%20city:u.s.a.*
&facet=true&facet.field=tags&facet.field=color
The query contains the high level "OR" query, consisting of 2 terms, 2
ranges and 1 prefix. It is designed to hit ~60-70% of all the docs
Here is the performance result:
#Threads     min       median         mean            75%         qps
   1         208.95ms  332.66ms    350.48ms     422.92ms     2.8
   2         188.68ms  338.09ms    339.22ms     402.15ms     5.9
   3         151.06ms  326.64ms    336.20ms     418.61ms     8.8
   4         125.13ms  332.90ms    332.18ms     396.14ms     12.0
If there is no  indexing process on background
The result is as follows for 2,6 mln docs:
#Threads     min     median          mean             75%         qps
   1         106.70ms  199.66ms    199.40ms     234.89ms     5.1
   2         128.61ms  199.12ms    201.81ms     229.89ms     9.9
   3         110.99ms  197.43ms    203.13ms     232.25ms     14.7
   4         90.24ms    201.46ms      200.46ms     227.75ms     19.9
   5         106.14ms  208.75ms    207.69ms     242.88ms     24.0
   6         103.75ms  208.91ms    211.23ms     238.60ms     28.3
   7         113.54ms  207.07ms    209.69ms     239.99ms     33.3
   8         117.32ms  216.38ms    224.74ms     258.74ms     35.5
I've got three questions so far:
1. In case of background indexing the latency is almost 2 times
higher, is there any way to overcome this?
2. How can we tune the Solr to get better results ?
3. What's in your opinion is the preferred type of queries that I can
use for the benchmark?

With many thanks,
Volodymyr


BTW here is the spec of my machine
RedHat 6.1 64bit
Intel XEON e5620 @2.40 GHz, 8 cores
63 GB RAM
Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message