Data-oriented design

In computing, data-oriented design is a program optimization approach motivated by efficient usage of the CPU cache, used in video game development.[1] The approach is to focus on the data layout, separating and sorting fields according to when they are needed, and to think about transformations of data. Proponents include Mike Acton[2] and Scott Meyers[3].

MotivesEdit

These methods became especially popular in the mid to late 2000s during the seventh generation of video game consoles that included PlayStation 3 (PS3) and Xbox 360. Historically, game consoles often have relatively weak central processing units (CPUs) compared to the top-of-line desktop computer counterparts. This is a design choice to devote more power and transistor budget to the graphics processing units (GPUs). For example, the 7th generation CPUs were not manufactured with modern out-of-order execution processors, but instead use in-order processors with high clock speeds and deep pipelines. In addition, most types of computing systems have main memory located hundreds of clock cycles away from the processing elements. Furthermore, as CPUs have become faster alongside a large increase in main memory capacity, there is massive data consumption that increases the likelihood of cache misses in the shared bus, otherwise known as Von Neumann bottlenecking. Consequently, locality of reference methods have been used to control performance, requiring improvement of memory access patterns to fix bottlenecking. Some of the software issues were also similar to those encountered on the Itanium, requiring loop unrolling for upfront scheduling.

Contrast with object orientationEdit

The claim is that traditional object-oriented programming (OOP) design principles result in poor data locality, more so if runtime polymorphism (dynamic dispatch) is used (which is especially problematic on some processors).[4][5] Although OOP does seem to "organise code around data", the practice is quite different. OOP is actually about organising source code around data types rather than physically grouping individual fields and arrays in an efficient format for access by specific functions. Moreover, it often hides layout details under abstraction layers, while a data-oriented programmer wants to consider this first and foremost.

Programming languagesEdit

The experimental programming language JAI being developed by Jonathan Blow has explicit support for data-oriented design that eschews the traditional OOP paradigm. This is facilitated by being able to transparently move fields between records without extensive source code changes to functions using them (or without extensive boilerplate code to enable this), and by adding direct support for structure of arrays (SoA) data layout.[6]

See alsoEdit

ReferencesEdit

  1. ^ "Data-oriented design" (PDF).
  2. ^ "CppCon 2014: Mike Acton "Data-Oriented Design and C++"".
  3. ^ "code::dive conference 2014 - Scott Meyers: Cpu Caches and Why You Care".
  4. ^ "What's wrong with Object-Oriented Design? Where's the harm in it?".describes the problems with virtual function calls, e.g., i-cache misses
  5. ^ "Data-oriented design - why you might be shooting yourself in the foot with OOP".
  6. ^ "Data-oriented demo:SOA,composition".Demonstration of data-oriented and SOA features in the JAI language, also explaining the motives.