Skip to main content

What is Shuffling in Spark

Shuffling in Spark is a mechanism that Re-Distributes the data across different executors or workers in the clusters. 

Why do we need to Re-Distribute the data? 

 A) Re-Distribution is needed when there is a need of increasing or decreasing the data partitions in the situations below:

  • When the partitions are not sufficient enough to process the data load in the cluster
  • When the partitions are too high in numbers that it creates task scheduling overhead and it becomes the bottleneck in the processing time.
Re-Distribution can also be achieved by executing the shuffling on existing distributed data collection like RDD, DataFrames, etc by using the "Repartition" and "Coalesce" APIs in Spark.

B) During Aggregation and Joins on data collection in Spark, all the data records belonging to aggregation or join should reside in the single partition and when the existing partitioning scheme doesn't satisfy this condition there is a need to re-distributing the data in input collections before performing the aggregations or joins.

Spark Shuffling is an expensive process as it is moving around data among different executors or workers in the cluster. Imagine, if you have 1000s of workers and have a huge volume of data, in this case, the Spark Shuffling can lead to expensive overheads.

Generally, Spark Shuffling involves
  • Disk I/O
  • Network I/O
  • Data Serialization & Deserialization
Some of the operations that might lead to the Shuffling are 

  • join
  • cogroup
  • groupWith
  • join
  • leftOuterJoin
  • rightOuterJoin
  • groupByKey
  • reduceByKey
  • repartition
  • coalesce
Data Engineers should always consider the possibility where they can avoid Shuffling in their process. For Example, they can choose to use reduceByKey instead of groupByKey due to the following reasons.

  • groupByKey shuffles all the data and therefore it is too slow.
  • reduceByKey shuffles only the sub-aggregation results in the partitions.

To summarize, Shuffling is one of the important operations of Spark that re-distributes the data to make the pipelines more efficient. But we are also aware of how expensive Shuffling could be especially when the number of workers in a cluster is too high or the volume of data is huge. So one should avoid the Shuffling in their pipelines. Also at the same time, one should also understand the concept clearly that could help them to design Fault-tolerant, Robust, and Reliable Data Pipelines.

Comments

Post a Comment

Popular posts from this blog

Building Scalable and Efficient Data Lakes with Apache Hudi

If you're looking to build a scalable and efficient data lake that can support both batch and real-time processing, Apache Hudi is a great tool to consider. In this blog post, we'll discuss what Apache Hudi is, how it works, and why it's a powerful tool for building data lakes. Apache Hudi is an open-source data management framework that provides several features to manage big data. It provides the ability to perform read and write operations on large datasets in real-time, while also supporting batch processing. With Hudi, you can also ensure data quality by performing data validation, data cleansing, and data profiling. One of the key advantages of Apache Hudi is its support for schema evolution. This means that as your data changes over time, Hudi can automatically update the schema of your data to accommodate these changes, without requiring any downtime or manual intervention. Another advantage of Hudi is its support for scalable and fault-tolerant data storage. Hudi p...
Post a Comment
Read more

Top 25 Data Engineer Interview Questions

In my last post  How to prepare for Data Engineer Interviews ,  I wrote about how one can prepare for the Data Engineer Interviews, and in this blog post, I am going to provide the  Top 25 Basic   data engineer interview questions  asked frequently and their brief answers. This is typically the first round of the Interview where the interviewer just wants to access whether you are aware of basic concepts or not and therefore you don't need to explain it in detail. Just a single statement would be sufficient. Let's get started Checkout the 5 Key Skills Data Engineers need in 2023 A. Programming  1. What is the Static method in Python? Static methods are the methods that are bound to the  Class  rather than the Class's Object. Thus, it can be called without creating objects of the class. We can just call it using the reference of the class. Also, all the objects of the class share only one copy of the static method. 2. What is a Decorator in Python?...
Post a Comment
Read more

What is CAP Theorem?

CAP Theorem states that a Distributed Database System can only have 2 out of 3 properties from Availability, Consistency, and Partition Tolerance . This means that every Big Data Engineer needs to do a trade-off between these three based on the use-case and Business requirements. It is very important for any Data engineer to understand the CAP Theorem and apply it when deciding the appropriate tools for the task in the hand. Let's discuss each of the properties in detail. 1. Availability   This condition states that every request (read/write) will get a response on Success or Failure. That means every node in the system must return a response in a reasonable amount of time. This could be only possible if the system remains operational all the time. Hence, the databases are time-independent as they should be available all the time. Therefore if any two records are added to the database we don't know which one was added first and the output could be either one of them. Now le...
Post a Comment
Read more