Elasticsearch does not support streaming in the sense, that you subscribe to a search query and as new documents come in, you will get any matching your query. There is a rather long discussion in one of the issues in the ES repository on GitHub concerning this (active for 4 years now) [1]
It does however provide you with an API to scroll through a large set of results, so you can retrieve them in chunks which fit your application. [2] That is supported by the hadoop connector via [3] which is used as far as I can tell in almost all of the different connectors (hive, cascading, mr, spark) Best regards, Martin [1] https://github.com/elastic/elasticsearch/issues/1242 [2] https://www.elastic.co/guide/en/elasticsearch/reference/current/search-request-scroll.html [3] https://github.com/elastic/elasticsearch-hadoop/blob/03c056142a5ab7422b81bb1f519fd67a9581405f/mr/src/main/java/org/elasticsearch/hadoop/rest/ScrollQuery.java Robert Metzger <rmetz...@apache.org> schrieb am Mi., 14. Okt. 2015 um 17:49 Uhr: > I think the Hadoop connector of ElasticSearch only works for batch jobs. > In my understanding, elasticsearch also allows "streaming" standing search > queries. > > On Wed, Oct 14, 2015 at 10:41 AM, santosh_rajaguru <sani...@gmail.com> > wrote: > > > Thanks flavio. > > > > > > > > -- > > View this message in context: > > > http://apache-flink-mailing-list-archive.1008284.n3.nabble.com/Apache-flink-with-Elastic-Search-tp8435p8445.html > > Sent from the Apache Flink Mailing List archive. mailing list archive at > > Nabble.com. > > >
