How to iteratively bring a Dataframe back to the driver in Spark - scala

Main question: How do you safely (without risking crashing due to OOM) iterate over every row (guaranteed every row) in a dataframe from the driver node in Spark? I need to control how big the data is as it comes back, operate on it, and discard it to retrieve the next batch (say 1000 rows at a time or something)
I am trying to safaely and iteratively bring the data in a potentially large Dataframe back to the driver program so that I may use the data to perform HTTP calls. I have been attempting use someDf.foreachPartition{makeApiCall(_)} and allowing the Executors to handle the calls. It works - but debugging and handling errors has proven to be pretty difficult when launching in prod envs, especially on failed calls.
I know there is someDf.collect() action, which brings ALL the data back to the driver all at once. However, this solution is not suggested, because if you have a very large DF, you risk crashing the driver.
Any suggestions?

if the data does not fit into memory, you could use something like :
df.toLocalIterator().forEachRemaining( row => {makeAPICall(row)})
but toLocalIterator has considerable overhead compared to collect
Or you can collect your dataframe batch-wise (which does essentially the same as toLocalIterator):
val partitions = df.rdd.partitions.map(_.index)
partitions.toStream.foreach(i => df.where(spark_partition_id() === lit(i)).collect().map(row => makeAPICall(row)))

It is a bad idea to bring all that data back to driver because driver is just 1 node and it will become the bottleneck. The scalability will be lost. If you had to do this then think twice if you really need a big data application? probably not.
dataframe.collect() is the best way to bring the data to driver and it will bring all data. The alternative is toLocalIterator which will bring data of the largest partition which can be big too. So this should be rarely used and for small amount of data only.
If you insist then you can write the output to a file or queue and read that file in a controlled manner. This will be partially scalable solution which I won't prefer.

Related

Is it possible to generate DataFrame rows from the context of a Spark Worker?

The fundamental problem is attempting to use spark to generate data but then work with the data internally. I.e., I have a program that does a thing, and it generates "rows" of data - can I leverage Spark to parallelize that work across the worker nodes, and have them each contribute back to the underlying store?
The reason I want to use Spark is that seems to be a very popular framework, and I know this request is a little outside of the defined range of functions Spark should offer. However, the alternatives of MapReduce or Storm are dreadfully old and there isn't much support anymore.
I have a feeling there has to be a way to do this, has anyone tried to utilize Spark in this way?
To be honest, I don't think adopting Spark just because it's popular is the right decision. Also, it's not obvious from the question why this problem would require a framework for distributed data processing (that comes along with a significant coordination overhead).
The key consideration should be how you are going to process the generated data in the next step. If it's all about dumping it immediately into a data store I would really discourage using Spark, especially if you don't have the necessary infrastructure (Spark cluster) at hand.
Instead, write a simple program that generates the data. Then run it on a modern resource scheduler such as Kubernetes and scale it out and run as many instances of it as necessary.
If you absolutely want to use Spark for this (and unnecessarily burn resources), it's not difficult. Create a distributed "seed" dataset / stream and simply flatMap that. Using flatMap you can generate as many new rows for each seed input row as you like (obviously limited by the available memory).

What happens to a Spark DataFrame used in Structured Streaming when its underlying data is updated at the source?

I have a use case where I am joining a streaming DataFrame with a static DataFrame. The static DataFrame is read from a parquet table (a directory containing parquet files).
This parquet data is updated by another process once a day.
My question is what would happen to my static DataFrame?
Would it update itself because of the lazy execution or is there some weird caching behavior that can prevent this?
Can the updation process make my code crash?
Would it be possible to force the DataFrame to update itself once a day in any way?
I don't have any code to share for this because I haven't written any yet, I am just exploring what the possibilities are. I am working with Spark 2.3.2
A big (set of) question(s).
I have not implemented all aspects myself (yet), but this is my understanding and one set of info from colleagues who performed an aspect that I found compelling and also logical. I note that there is not enough info out there on this topic.
So, if you have a JOIN (streaming --> static), then:
If standard coding practices as per Databricks applied and .cache is applied, the SparkStructuredStreamingProgram will read in static source only once, and no changes seen on subsequent processing cycles and no program failure.
If standard coding practices as per Databricks applied and caching NOT used, the SparkStructuredStreamingProgram will read in static source every loop, and all changes will be seen on subsequent processing cycles hencewith.
But, JOINing for LARGE static sources not a good idea. If large dataset evident, use Hbase, or some other other key value store, with mapPartitions if volitatile or non-volatile. This is more difficult though. It was done by an airline company I worked at and was no easy task the data engineer, designer told me. Indeed, it is not that easy.
So, we can say that updates to static source will not cause any crash.
"...Would it be possible to force the DataFrame to update itself once a day in any way..." I have not seen any approach like this in the docs or here on SO. You could make the static source a dataframe using var, and use a counter on the driver. As the micro batch physical plan is evaluated and genned every time, no issue with broadcast join aspects or optimization is my take. Whether this is the most elegant, is debatable - and is not my preference.
If your data is small enough, the alternative is to read using a JOIN and thus perform the look up, via the use of the primary key augmented with some max value in a
technical column that is added to the key to make the primary key a
compound primary key - and that the data is updated in the background with a new set of data, thus not overwritten. Easiest
in my view if you know the data is volatile and the data is small. Versioning means others may still read older data. That is why I state this, it may be a shared resource.
The final say for me is that I would NOT want to JOIN with the latest info if the static source is large - e.g. some Chinese
companies have 100M customers! In this case I would use a KV store as
LKP using mapPartitions as opposed to JOIN. See
https://medium.com/#anchitsharma1994/hbase-lookup-in-spark-streaming-acafe28cb0dc
that provides some insights. Also, this is old but still applicable
source of information:
https://blog.codecentric.de/en/2017/07/lookup-additional-data-in-spark-streaming/.
Both are good reads. But requires some experience and to see the
forest for the trees.

Convert a JavaPairRDD into list without collect() [duplicate]

We know that if we need to convert RDD to a list, then we should use collect(). but this function puts a lot of stress on the driver (as it brings all the data from different executors to the driver) which causes performance degradation or worse (whole application may fail).
Is there any other way to convert RDD into any of the java util collection without using collect() or collectAsMap() etc which does not cause performance degrade?
Basically in current scenario where we deal with huge amount of data in batch or stream data processing, APIs like collect() and collectAsMap() has become completely useless in a real project with real amount of data. We can use it in demo code, but that's all there to use for these APIs. So why to have an API which we can not even use (Or am I missing something).
Can there be a better way to achieve the same result through some other method or can we implement collect() and collectAsMap() in a more effective way other that just calling
List<String> myList= RDD.collect.toList (which effects performance)
I looked up to google but could not find anything which can be effective. Please help if someone has got a better approach.
Is there any other way to convert RDD into any of the java util collection without using collect() or collectAsMap() etc which does not cause performance degrade?
No, and there can't be. And if there were such a way, collect would be implemented using it in the first place.
Well, technically you could implement List interface on top of RDD (or most of it?), but that would be a bad idea and quite pointless.
So why to have an API which we can not even use (Or am I missing something).
collect is intended to be used for cases where only large RDDs are inputs or intermediate results, and the output is small enough. If that's not your case, use foreach or other actions instead.
As you want to collect the Data in a Java Collection, the data has to collect on single JVM as the java collections won't be distributed. There is no way to get all data in collection by not getting data. The interpretation of problem space is wrong.
collect and similar are not meant to be used in normal spark code. They are useful for things like debugging, testing, and in some cases when working with small datasets.
You need to keep your data inside of the rdd, and use rdd transformations and actions without ever taking the data out. Methods like collect which pull you data out of spark and onto your driver defeat the purpose and undo any advantage that spark might be providing since now you're processing all of your data on a single machine anyway.

What's the difference between collect and count actions?

When I was coding scala program in local-mode spark, the code is similar to RDD.map(x => function()).collect, there is no log output in the console for a long time and I guess it stucks. So I change the action collect into count, the whole execution completed soon. Plus, there is little records produced by the map phase to be collected by collect, so problem can't be caused by the network trans when sending back the result to the driver.
Can anyone know the reason or has run into the similar problem?
The method count sums the number of entries of the RDD for every partition, and it returns an integer consisting in that number, hence the data transfer is minimal; in the other hand, the method collect as its name says brings all the entries to the driver as a List therefore if there isn't enough memory you may get several exceptions (this is why it's not recommended to do a collect if you aren't sure that it will fit in your driver, there are normally faster alternatives like take, which also triggers lazy transformations), besides it requires to transfer much more data.

Apache spark streaming - cache dataset for joining

I'm considering using Apache Spark streaming for some real-time work but I'm not sure how to cache a dataset for use in a join/lookup.
The main input will be json records coming from Kafka that contain an Id, I want to translate that id into a name using a lookup dataset. The lookup dataset resides in Mongo Db but I want to be able to cache it inside the spark process as the dataset changes very rarely (once every couple of hours) so I don't want to hit mongo for every input record or reload all the records in every spark batch but I need to be able to update the data held in spark periodically (e.g. every 2 hours).
What is the best way to do this?
Thanks.
I've thought long and hard about this myself. In particular I've wondered is it possible to actually implement a database DB in Spark of sorts.
Well the answer is kind of yes. First you want a program that first caches the main data set into memory, then every couple of hours does an optimized join-with-tiny to update the main data set. Now apparently Spark will have a method that does a join-with-tiny (maybe it's already out in 1.0.0 - my stack is stuck on 0.9.0 until CDH 5.1.0 is out).
Anyway, you can manually implement a join-with-tiny, by taking the periodic bi-hourly dataset and turning it into a HashMap then broadcasting it as a broadcast variable. What this means is that the HashMap will be copied, but only once per node (compare this with just referencing the Map - it would be copied once per task - a much greater cost). Then you take your main dataset and add on the new records using the broadcasted map. You can then periodically (nightly) save to hdfs or something.
So here is some scruffy pseudo code to elucidate:
var mainDataSet: RDD[KeyType, DataType] = sc.textFile("/path/to/main/dataset")
.map(parseJsonAndGetTheKey).cache()
everyTwoHoursDo {
val newData: Map[KeyType, DataType] = sc.textFile("/path/to/last/two/hours")
.map(parseJsonAndGetTheKey).toarray().toMap
broadcast(newData)
val mainDataSetNew =
mainDataSet.map((key, oldValue) => (key,
newData.get(key).map(newDataValue =>
update(oldValue, newDataValue))
.getOrElse(oldValue)))
.cache()
mainDataSetNew.someAction() // to force execution
mainDataSet.unpersist()
mainDataSet = mainDataSetNew
}
I've also thought that you could be very clever and use a custom partioner with your own custom index, and then use a custom way of updating the partitions so that each partition itself holds a submap. Then you can skip updating partitions that you know won't hold any keys that occur in the newData, and also optimize the updating process.
I personally think this is a really cool idea, and the nice thing is your dataset is already ready in memory for some analysis / machine learning. The down side is your kinda reinventing the wheel a bit. It might be a better idea to look at using Cassandra as Datastax is partnering with Databricks (people who make Spark) and might end up supporting some kind of thing like this out of box.
Further reading:
http://spark.apache.org/docs/latest/programming-guide.html#broadcast-variables
http://www.datastax.com/2014/06/datastax-unveils-dse-45-the-future-of-the-distributed-database-management-system
Here is a fairly simple work-flow:
For each batch of data:
Convert the batch of JSON data to a DataFrame (b_df).
Read the lookup dataset from MongoDB as a DataFrame (m_df). Then cache, m_df.cache()
Join the data using b_df.join(m_df, "join_field")
Perform your required aggregation and then write to a data source.