DProf is a statistical profiler that helps programmers understand cache miss costs by attributing misses to data types instead of code.

Publications

Locating Cache Performance Bottlenecks Using Data Profiling Abstract  BibTeX  PDF
Aleksey Pesterev, Nickolai Zeldovich, and Robert T. Morris. In the Proceedings of the ACM EuroSys Conference (EuroSys 2010), Paris, France, April 2010.

Effective use of CPU data caches is critical to good performance, but poor cache use patterns are often hard to spot using existing execution profiling tools. Typical profilers attribute costs to specific code locations. The costs due to frequent cache misses on a given piece of data, however, may be spread over instructions throughout the application. The resulting individually small costs at a large number of instructions can easily appear insignificant in a code profiler's output.

DProf helps programmers understand cache miss costs by attributing misses to data types instead of code. Associating cache misses with data helps programmers locate data structures that experience misses in many places in the application's code. DProf introduces a number of new views of cache miss data, including a data profile, which reports the data types with the most cache misses, and a data flow graph, which summarizes how objects of a given type are accessed throughout their lifetime, and which accesses incur expensive cross-CPU cache loads. We present two case studies of using DProf to find and fix cache performance bottlenecks in Linux. The improvements provide a 16--57% throughput improvement on a range of memcached and Apache workloads.

Software

Beware that the code is rough and it may not work on your machine. This code requires an AMD processor with IBS support.

$ git clone https://pdos-csail-mit-edu.ezproxy.canberra.edu.au/dprof/dprof-linux.git

This is the Linux kernel to be used with DProf. The repository is composed of multiple branches; each branch contains changes to make data collection possible. The master branch is a merge of all branches. The .config we use is also available from the mosbench page.

$ git clone https://pdos-csail-mit-edu.ezproxy.canberra.edu.au/dprof/slab_data_profile.git

This module allows DProf to collect IBS samples annotated with type information.

$ git clone https://pdos-csail-mit-edu.ezproxy.canberra.edu.au/dprof/object_flow_profile.git

This module allows DProf to collect the set of instructions that access a particulars kernel object.

$ git clone https://pdos-csail-mit-edu.ezproxy.canberra.edu.au/dprof/dprof.git

This is a set of Ruby scripts to aggregate the collected data.