Hm, that's a pretty big question. For one, open-source engines are
cheaper.
But maybe you have specific requirements or features you want to inquire
about?
Probably not the best place here to ask. maybe quora.
Apart from the cost and support, what about performance?
On Thursday, May 23, 2013 3:29:11 AM UTC+8, Otis Gospodnetic wrote:
Hi Paul,
Hm, that's a pretty big question. For one, open-source engines are
cheaper.
But maybe you have specific requirements or features you want to inquire
about?
Short answer: I don't know.
Medium answer: I'm sure each vendor would claim superiority or, if losing
in perf, would bring up functionality their tool has that others do not
True answer: you would really have to test it yourself. There are a
million ways to do comparative benchmarks and lots of different use cases,
that it's impossible to test them all and provide truthful reports.
What I can say though is that OSS search engines like ES and Solr can be very fast if you know how to use them. For example, Sematext had a
client recently whose initial queries against a 500K doc index took 90
seconds. We changed the structure of the index in a way that may have
seemed crazy to some and ended up with 10x "taller" index that was
returning results in just a few seconds on a single machine. So since
open-source engines are free to use, you should simply try them out and see
if they are fast enough for you, for your use case. I'll bet my left
pinkie that you'll find they'll be plenty fast. And if they are not, you
can get help on this ML or from companies that provide ES consulting
services.
On Thursday, May 23, 2013 5:10:09 AM UTC-4, Paul wrote:
Hi Otis
Probably not the best place here to ask. maybe quora.
Apart from the cost and support, what about performance?
On Thursday, May 23, 2013 3:29:11 AM UTC+8, Otis Gospodnetic wrote:
Hi Paul,
Hm, that's a pretty big question. For one, open-source engines are
cheaper.
But maybe you have specific requirements or features you want to inquire
about?
Otis,
I know this is old, but what do you mean by "taller"?
On Friday, May 24, 2013 12:28:17 PM UTC-4, Otis Gospodnetic wrote:
Hi,
Short answer: I don't know.
Medium answer: I'm sure each vendor would claim superiority or, if losing
in perf, would bring up functionality their tool has that others do not
True answer: you would really have to test it yourself. There are a
million ways to do comparative benchmarks and lots of different use cases,
that it's impossible to test them all and provide truthful reports.
What I can say though is that OSS search engines like ES and Solr can be very fast if you know how to use them. For example, Sematext had a
client recently whose initial queries against a 500K doc index took 90
seconds. We changed the structure of the index in a way that may have
seemed crazy to some and ended up with 10x "taller" index that was
returning results in just a few seconds on a single machine. So since
open-source engines are free to use, you should simply try them out and see
if they are fast enough for you, for your use case. I'll bet my left
pinkie that you'll find they'll be plenty fast. And if they are not, you
can get help on this ML or from companies that provide ES consulting
services.
On Thursday, May 23, 2013 5:10:09 AM UTC-4, Paul wrote:
Hi Otis
Probably not the best place here to ask. maybe quora.
Apart from the cost and support, what about performance?
On Thursday, May 23, 2013 3:29:11 AM UTC+8, Otis Gospodnetic wrote:
Hi Paul,
Hm, that's a pretty big question. For one, open-source engines are
cheaper.
But maybe you have specific requirements or features you want to inquire
about?
On Wednesday, August 13, 2014 6:09:07 PM UTC+2, Shawn Johnson wrote:
Otis,
I know this is old, but what do you mean by "taller"?
On Friday, May 24, 2013 12:28:17 PM UTC-4, Otis Gospodnetic wrote:
Hi,
Short answer: I don't know.
Medium answer: I'm sure each vendor would claim superiority or, if losing
in perf, would bring up functionality their tool has that others do not
True answer: you would really have to test it yourself. There are a
million ways to do comparative benchmarks and lots of different use cases,
that it's impossible to test them all and provide truthful reports.
What I can say though is that OSS search engines like ES and Solr can be very fast if you know how to use them. For example, Sematext had a
client recently whose initial queries against a 500K doc index took 90
seconds. We changed the structure of the index in a way that may have
seemed crazy to some and ended up with 10x "taller" index that was
returning results in just a few seconds on a single machine. So since
open-source engines are free to use, you should simply try them out and see
if they are fast enough for you, for your use case. I'll bet my left
pinkie that you'll find they'll be plenty fast. And if they are not, you
can get help on this ML or from companies that provide ES consulting
services.
On Thursday, May 23, 2013 5:10:09 AM UTC-4, Paul wrote:
Hi Otis
Probably not the best place here to ask. maybe quora.
Apart from the cost and support, what about performance?
On Thursday, May 23, 2013 3:29:11 AM UTC+8, Otis Gospodnetic wrote:
Hi Paul,
Hm, that's a pretty big question. For one, open-source engines are
cheaper.
But maybe you have specific requirements or features you want to
inquire about?
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.