# KTable connector?

**URL:** <https://community.dremio.com/t/ktable-connector/1124>\
**Category:** Uncategorized\
**Created:** [May 13, 2018, 8:08am UTC](https://community.dremio.com/t/ktable-connector/1124 "2018-05-13T08:08:10Z")\
**Posts on this page:** 7\
**Page:** 1

<div class="post-metadata">

**Author:** ![xmehaut](https://sea2.discourse-cdn.com/flex020/user_avatar/community.dremio.com/xmehaut/32/600_2.png) [@xmehaut](https://community.dremio.com/u/xmehaut)\
**Post date:** [May 13, 2018, 8:08am UTC](https://community.dremio.com/t/ktable-connector/1124/1 "2018-05-13T08:08:10Z")

</div>

Hello,  
is it planned to use Kafka KTable as source for dremio? It could be great for building an IOT plaform with Kafka and Dremio where Kafka is used as a streaming datastore and Dremio as a query engine and data governance tool. No more need of hadoop then.

---

<div class="post-metadata">

**Author:** ![anthony](https://sea2.discourse-cdn.com/flex020/user_avatar/community.dremio.com/anthony/32/670_2.png) [@anthony](https://community.dremio.com/u/anthony)\
**Post date:** [May 13, 2018, 3:02pm UTC](https://community.dremio.com/t/ktable-connector/1124/2 "2018-05-13T15:02:29Z")

</div>

Kafka is currently not supported as a source but is on our roadmap.  
The best option right now is to output the stream to a file system like HDFS/S3/etc then access that as a source.

---

<div class="post-metadata">

**Author:** ![xmehaut](https://sea2.discourse-cdn.com/flex020/user_avatar/community.dremio.com/xmehaut/32/600_2.png) [@xmehaut](https://community.dremio.com/u/xmehaut)\
**Post date:** [May 13, 2018, 3:49pm UTC](https://community.dremio.com/t/ktable-connector/1124/3 "2018-05-13T15:49:11Z")

</div>

thanks for the answer ; the solution given is what we would like to avoid actually, because we need a second cluster for storing data, when we have already a datastore with kafka.

Other wise, is a druid connector planned?

---

<div class="post-metadata">

**Author:** ![kelly](https://sea2.discourse-cdn.com/flex020/user_avatar/community.dremio.com/kelly/32/11_2.png) [@kelly](https://community.dremio.com/u/kelly)\
**Post date:** [May 13, 2018, 4:49pm UTC](https://community.dremio.com/t/ktable-connector/1124/4 "2018-05-13T16:49:10Z")

</div>

Druid as a source is something we are watching to guage community interest. It is not currently scheduled.

Hopefully a little later this year we will have an SDK so people can start to build their own connectors and contribute those back to the project when that makes sense.

---

<div class="post-metadata">

**Author:** ![xmehaut](https://sea2.discourse-cdn.com/flex020/user_avatar/community.dremio.com/xmehaut/32/600_2.png) [@xmehaut](https://community.dremio.com/u/xmehaut)\
**Post date:** [May 13, 2018, 4:52pm UTC](https://community.dremio.com/t/ktable-connector/1124/5 "2018-05-13T16:52:41Z")

</div>

Nice…

on another subject, currently the means to access dremio by spark is jdbc… could not be a specific connector?

---

<div class="post-metadata">

**Author:** ![kelly](https://sea2.discourse-cdn.com/flex020/user_avatar/community.dremio.com/kelly/32/11_2.png) [@kelly](https://community.dremio.com/u/kelly)\
**Post date:** [May 13, 2018, 5:06pm UTC](https://community.dremio.com/t/ktable-connector/1124/6 "2018-05-13T17:06:19Z")

</div>

That’s correct, JDBC is the best option right now probably.

We have so many things we want to build, and we are hard at work to get them released as quickly as possible. Unfortunately, there are only so many hours in the day. 🙂

---

<div class="post-metadata">

**Author:** ![xmehaut](https://sea2.discourse-cdn.com/flex020/user_avatar/community.dremio.com/xmehaut/32/600_2.png) [@xmehaut](https://community.dremio.com/u/xmehaut)\
**Post date:** [May 13, 2018, 5:17pm UTC](https://community.dremio.com/t/ktable-connector/1124/7 "2018-05-13T17:17:51Z")

</div>

don’t worry 🙂  
it was just a question to know if it was the best way intrinsequally

regards