Thanks, it seems good, though a little hack.
And here is another question. updateByKey compute on all the data from the
beginning, but in many situation, we just need to update the coming data.
This could be a big improve on speed and resource. Would this to be support
in the future?
Shixiong Zhu
You can create connection like this:
val updateFunc = (iterator: Iterator[(String, Seq[Int], Option[Int])])
=> {
val dbConnection = create a db connection
iterator.flatMap { case (key, values, stateOption) =>
if (values.isEmpty) {
// don't access database
It seems like a work around. But I don't know how to get the database
connection from the working nodes.
Shixiong Zhu 于2015年9月24日周四 下午5:37写道：
> Could you write your update func like this?
>
> val updateFunc = (iterator: Iterator[(String, Seq[Int], Option[Int])])
> => {
> iterator.flatMa
Could you write your update func like this?
val updateFunc = (iterator: Iterator[(String, Seq[Int], Option[Int])])
=> {
iterator.flatMap { case (key, values, stateOption) =>
if (values.isEmpty) {
// don't access database
} else {
// update to new state
Data that are not updated should be saved earlier: while the data added to
the DStream at the first time, it should be considered as updated. So save
the same data again is a waste.
What are the community is doing? Is there any doc or discussion that I can
look for? Thanks.
Shixiong Zhu 于2015年9
For data that are not updated, where do you save? Or do you only want to
avoid accessing database for those that are not updated?
Besides, the community is working on optimizing "updateStateBykey"'s
performance. Hope it will be delivered soon.
Best Regards,
Shixiong Zhu
2015-09-24 13:45 GMT+08:
I've read the source code and it seems to be impossible, but I'd like to
confirm it.
It is a very useful feature. For example, I need to store the state of
DStream into my database, in order to recovery them from next redeploy. But
I only need to save the updated ones. Save all keys into database
