Looking for something else? Check the categories of Data processing:
Apache Beam Apache Flink Apache Spark Apache Spark GraphFrames Apache Spark GraphX Apache Spark SQL Apache Spark Streaming Apache Spark Structured Streaming PySpark
If not, below you can find all articles belonging to Data processing.
In November 2018 bithw1 pointed out to me a feature that I haven't used yet in Apache Spark - custom optimization. After some months consacred to learning Apache Spark GraphX, I finally found a moment to explore it. This post begins a new series about Apache Spark customization and it covers the basics, i.e. the 2 available methods to add the custom optimizations.
In the previous post in GraphFrames category I mentioned the motifs finding feature and promised to write a separate post about it. After several weeks dedicated to the Apache Spark 2.4.0 features, I finally managed to find some time to explore the motifs finding in GraphFrames.
Some time ago I was asked by Sunil whether it was possible to load the initial state in Apache Spark Structured Streaming like in DStream-based API. Since the response was not obvious, I decided to investigate and share the findings through this post.
In previous posts about memory in Apache Spark, I've been exploring memory behavior of Apache Spark when the input files are much bigger than the allocated memory. After that it's a good moment to sum up that in the post dedicated to classes involved in memory using tasks.
For a long time, I've wanted to make a small real-time data visualization application with the use of websockets and some fancy JavaScript visualization framework. And the moment went when I was preparing the execution schemas to illustrate distributed graph algorithms covered in Graph algorithms in distributed world - part 1 post. I used there static images combined together but it was quite painful. Because of that, I decided to check whether it's possible to do in a more programmatic way.
When I first heard about the foreachBatch feature, I thought that it was the implementation of foreachPartition in the Structured Streaming module. However, after some analysis I saw how I was wrong because this new feature addresses other but also important problems. You will find more .
Data-driven systems continuously change. We moved from static, batch-oriented daily processing jobs to real-time streaming-based pipelines running all the time. Nowadays, the workflows have more and more AI compontents. Apache Spark tries to stay in the movement and in the new release proposes the implementation of the barrier execution mode as a new way to schedule tasks.
The Project Tungsten revolutionized Apache Spark ecosystem. Thanks to the new row-based data structure the jobs became more performant and easier to create. This revolution first affected the batch processing and later the streaming one. As of writing the following article, the graph processing is still not impacted but hopefully GraphFrames project can change this.
Apache Avro became one of the serialization standards, among others because of its use in Apache Kafka's schema registry. Previously to work with Avro files with Apache Spark we needed Databrick's external package. But it's no longer the case starting from 2.4.0 release where Avro became first-class citizen data source.
One of important characteristics of distributed graph processing which makes it different from classical Map/Reduce approach is the iterative nature of many algorithms. Pregel is one of the computation models that supports such kind of processing very well, while Apache Spark GraphX comes with its own Pregel implementation.
The series about the features introduced in Apache Spark 2.4.0 continues. Today's post will cover higher-order functions that you may know from elsewhere.
Until now we've been working only with in-memory graphs. However, Apache Spark GraphX provides a much more convenient and prod-ready methods to load and save them. And this post will try to show them.
The series about Apache Spark 2.4.0 features continues. After last week's discovery of bucket pruning, it's time to switch to Structured Streaming module and see its major evolution.
Previously we've learned about the vertices and edges representations in Apache Spark GraphX. At this moment to not introduce too many new concepts at once, we deliberately omitted the discovery of edges partitioning. Luckily, a new week comes and it lets us discuss that.
This post begins a new series dedicated to Apache Spark 2.4.0 features. The first covered topic will be bucket pruning.
Schemas are one of the key parts of Apache Spark SQL and its distinction point with old RDD-based API. When we deal with data coming from a structured data source as a relational database or schema-based file formats, we can let the framework to resolve the schema for us. But the things complicate when we're working with semi-structured data as JSON and we must define the schema by hand.
After last week's discovery of VertexRDD we have still one graph-composing item to explain - EdgeRDD. After all, the graph is about the relationships this RDD guarantees the links between vertices.
The code generated by Apache Spark for all the queries defined with higher-level concepts as SQL queries is the key to understand the processing logic performance. This post, started after a discussion on my Github, tries to explain some of the basics of code generation workflow.
After last week's global overview of graph representation in GraphX module, it's time to go a little bit deeper and analyze the 2 main components of graphs: vertices and edges. We'll begin here with the former ones.
Some weeks ago I've written a post about files with long lines impact on RDD-based processing. The post revealed the difficulty to process such files because of OOM errors. In this post I wanted to check how does it apply to Datasets.
