Approximation Accuracy in Cache Behavior Prediction

Approximation Accuracy in Cache Behavior Prediction

Memory transfers are the performance bottleneck of many applications due to poor data locality and limited memory bandwidth. Code refactoring for better data locality can improve cache behavior, leading to significant performance boosts. Reuse distance, a measurure of data locality, is useful in identification and optimization of hot code regions exhibiting poor data locality.

Benefits

Successful completion of the bachelor's thesis enables you to enter responsible positions, for example, in the areas of artificial intelligence, gaming and high-frequency trading. The ability to systematically identify and resolve performance bottlenecks is a highly sought-after skill.

Performance matters! During this thesis you can:

Motivation and Background

Reuse distance is defined as the number of unique memory locations referenced between a pair of references to the same memory location. On the granularity of cache lines, reuse distance can model spatial and temporal locality to assess cache behavior of applications. Assuming a fully associative cache with least recently used (LRU) replacement policy, predicting cache behavior with reuse distance is exact.

One of reuse distance's drawbacks is its high computational effort. To address this issue, recently, lightweight methods [1,2,3] combining hardware debug and profiling facilities with statistical approaches [4] have been developed. These methods measure reuse time, the total number of memory locations referenced between a pair of references to the same memory location, instead of reuse distance, and construct reuse distances from statistical considerations based on a Bernoulli process. Their approach, however, is based on the assumption that each memory location is referenced with equal probability. It is questionable whether this assumption holds, especially applied to highly irregular applications, such as sparse matrix-vector multiplication.

Goals and Tasks

In this thesis, you will explore the accuracy of cache behavior prediction with reuse distance in irregular applications. As an example application, you will use sequential and parallel sparse matrix-vector multiplications (SpMV), a ubiquitous kernel, for instance in simulations and graph algorithms.

In the context of this thesis, you will:

  1. Exact reuse distance
  2. Statistical methods based on memory instrumentation
  3. Lightweight methods based on statistical profiling
  4. Cachegrind (a Valgrind tool)

Additionally to accuracy of these methods, you will compare the induced runtime overhead.

Prerequisites

Starting Point Literature

Organization


Last Change: Fri, 07 Mar 2025 17:52:39 +0100 - Viewed on: Sun, 09 Mar 2025 23:48:47 +0100
Copyright © MNM-Team http://www.mnm-team.org - Impressum / Legal Info  - Datenschutz / Privacy