A row group is a unit of work for reading from Parquet that cannot be split into smaller parts, and you expect that the number of tasks created by Spark is no more than the total number of row groups in your Parquet data source.
But Spark still can create much more tasks than the number of row groups. Let’s see how this is possible.
A Parquet file contains MIN/MAX statistics for every column for every row group that allows Spark applications to skip reading unnecessary data chunks depending on the query predicate. Let’s see how this works with LIKE pattern matching filter.
For my tests I will use a Parquet file with 4 row groups and the following MIN/MAX statistics for product column:
It is quite common today to convert incoming JSON data into Parquet format to improve the performance of analytical queries.
When JSON data has an arbitrary schema i.e. different records can contain different key-value pairs, it is common to parse such JSON payloads into a map column in Parquet.
How is it stored? What read performance can you expect? Will json_map["key"] read only data for key or the entire JSON?
It is quite common to have a streaming Flink application that reads incoming data and puts them into Parquet files with low latency (a couple of minutes) for analysts to be able to run both near-realtime and historical ad-hoc analysis mostly using SQL queries.
But let’s review write patterns and problems that can appear for such applications at scale.
Amazon S3 is highly scalable distributed system that can handle extremely large volumes of data, can adapt to an increasing workload and provide quite good performance as a file storage.
But sometimes you have to tweak it to run faster that can be especially important for latency-sensitive applications.
Parquet is one of the most popular columnar file formats used in many tools including Apache Hive, Spark, Presto, Flink and many others.
For tuning Parquet file writes for various workloads and scenarios let’s see how the Parquet writer works in detail (as of Parquet 1.10 but most concepts apply to later versions as well).
Most applications writing data into S3 use the S3 multipart upload API to upload data in parts. First, you initiate the load, then upload parts and finally complete the multipart upload.
Let’s see how this operation is reflected in the S3 access log. My application uploaded the file data.gz into S3, and I can view it as follows:
One of our Flink streaming jobs had significant variance in the time spent on writing files to S3 by the same Task Manager process.
What settings do you need to check first?
I have a Spark job that transforms incoming data from compressed text files into Parquet format and loads them into a daily partition of a Hive table. This is a typical job in a data lake, it is quite simple but in my case it was very slow.
Initially it took about 4 hours to convert ~2,100 input .gz files (~1.9 TB of data) into Parquet, while the actual Spark job took just 38 minutes to run and the remaining time was spent on loading data into a Hive partition.
Let’s see what is the reason of such behavior and how we can improve the performance.
Traditional data warehouses require you to explicitly specify partition columns for tables using the PARTITION BY clause. There is no PARTITION BY clause in the CREATE TABLE statement in Snowflake although it still heavily relies on partitions.
I already wrote about partitions in Snowflake (see, MIN/MAX Functions and Partition Pruning in Snowflake) but in this article I am going to investigate some more details.