About this book

This book is a community-driven effort to provide a document describing how to write idiomatically fast Haskell by providing insight into how Haskell behaves under the hood, describing various techniques and tools used to identify performance issues, categorizing common performance issues, and demonstrate how to tackle performance problems step-by-step using case studies.

Our toolbox

In this section we'll list the tools we will use throughout the book, give a short description of them and where to get them.

GHC

GHC comes with a time and space profiling system. The output of that system is used by other tools to provide further insight than what GHC offers by default.

profiteur

An interactive treemap visualiser for GHC .prof files.

profiterole

Alternative textual views of the profiling data generated by GHC.

hp2ps

A tool used to convert a .hp file produced by GHC into a PostScript graph of heap profile. This tool is bundled with GHC.

eventlog2html

A tool to visualise eventlogs. In particular, it creates interactive charts for the heap profiling information included in the eventlog.

ghc-debug

A set of libraries which allow you to inspect the heap of a running Haskell program from an external debugger.

ghc-event-analyze

A profile tool that uses GHC's eventlog system. It is useful for profiling code when GHC's normal profiling mode is not available, or when using profiling mode would perturb the code too much. It is also useful when you want time-profiling information with a breakdown over time rather than totals for the whole run.

What makes programs go fast?

In this chapter we'll explore how Haskell behaves under the hood.

We'll take a look at Haskell's evaluation semantics, laziness/strictness and how to control it, how to represent data wrt performance, allocation and its affect on performance, how various GHC affect performance, and more.