Apache Avro is a serialization system. Avro supports rich data structures, a compact binary encoding, and a container file for sequences of Avro data (often referred to as "Avro data files"). Avro is designed to be language-independent and there are several language bindings for it, including Java, C, C++, Python, and Ruby.
Avro does not rely on generated code, which means that processing data imported from Flume or Sqoop is simpler than using Hadoop Writables in Sequence Files, where you have to take care that the generated classes are on the processing job's classpath. Furthermore, Pig and Hive cannot easily process Sequence Files with custom Writables, so users often revert to using text, which has disadvantages from a compactness and compressibility point of view (compressed text is not generally splittable, making it difficult to process efficiently using MapReduce).
All components in CDH4 that produce or consume files support Avro data files as a file format. But bear in mind that because uniform Avro support is new, there may be some rough edges or missing features.
|<< Previous: Using the HttpFS Server with curl||Next: Avro Data Files >>|