Although a little out of date, there is a website dedicated to this:
Redshift looks to be order-of-magnitude faster than Impala or Shark in all the test. Does this mean that once RedShift supports user-defined functions, there is no competing solution that is any match? (Unless you want to avoid using the cloud)
Rcfile or parquet would be a more interesting benchmark.
In the "What's next?" section, they say they want to re-do the Impala tests using Parquet, which is a columnar format based on the Dremel whitepaper (http://parquet.io/).
Parquet was a joint effort between Cloudera and Twitter, and now it's being developed by many other companies. You can use it with Hive, Pig, MapReduce, Cascading, Crunch and I think Apache Drill's first milestone has adopted it as a columnar format as well. Parquet also allows you to use your Avro or Thrift schema (soon Protobuffs) to write Parquet data, too.
It's a separate project in the ecosystem and has its own roadmap (https://github.com/Parquet/parquet-mr).
"Here from HackerNews? This was originally posted several months ago. Check back in two weeks for an updated benchmark including newer versions of Hive, Impala, and Shark."