Just pushed support for bulk operations, see more here: http://github.com/elasticsearch/elasticsearch/issues/issue/371(pay
attention to the body format). I have been slowly enabling
elasticsearch over the past few versions to support it, and finally came up
with a performant REST format that I am happy with.
Some very initial numbers. Using the Java API, on a two nodes cluster with
a single index with 2 shards with replica all on local machine. Single
threaded client using the single index API gets around 3k docs per second.
Using the bulk API with 100 batch size, it fires up to 22k docs per second
(single thread) Would love to hear numbers from users on this.
Just pushed support for bulk operations, see more here: Bulk API · Issue #371 · elastic/elasticsearch · GitHub (pay attention to the body format). I have been slowly enabling elasticsearch over the past few versions to support it, and finally came up with a performant REST format that I am happy with.
Some very initial numbers. Using the Java API, on a two nodes cluster with a single index with 2 shards with replica all on local machine. Single threaded client using the single index API gets around 3k docs per second. Using the bulk API with 100 batch size, it fires up to 22k docs per second (single thread) Would love to hear numbers from users on this.
Just pushed support for bulk operations, see more here: Bulk API · Issue #371 · elastic/elasticsearch · GitHub(pay attention to the body format). I have been slowly enabling
elasticsearch over the past few versions to support it, and finally came up
with a performant REST format that I am happy with.
Some very initial numbers. Using the Java API, on a two nodes cluster
with a single index with 2 shards with replica all on local machine.
Single threaded client using the single index API gets around 3k docs per
second. Using the bulk API with 100 batch size, it fires up to 22k docs per
second (single thread) Would love to hear numbers from users on this.
Just pushed support for bulk operations, see more here: Bulk API · Issue #371 · elastic/elasticsearch · GitHub(pay attention to the body format). I have been slowly enabling
elasticsearch over the past few versions to support it, and finally came up
with a performant REST format that I am happy with.
Some very initial numbers. Using the Java API, on a two nodes cluster
with a single index with 2 shards with replica all on local machine.
Single threaded client using the single index API gets around 3k docs per
second. Using the bulk API with 100 batch size, it fires up to 22k docs per
second (single thread) Would love to hear numbers from users on this.
Just pushed support for bulk operations, see more here: Bulk API · Issue #371 · elastic/elasticsearch · GitHub(pay attention to the body format). I have been slowly enabling
elasticsearch over the past few versions to support it, and finally came up
with a performant REST format that I am happy with.
Some very initial numbers. Using the Java API, on a two nodes cluster
with a single index with 2 shards with replica all on local machine.
Single threaded client using the single index API gets around 3k docs per
second. Using the bulk API with 100 batch size, it fires up to 22k docs per
second (single thread) Would love to hear numbers from users on this.
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.