Find file
Fetching contributors…
Cannot retrieve contributors at this time
89 lines (61 sloc) 1.78 KB


Let's run a function. Unfortunately, it does something very complex, so it takes a long time to run:

>> tic; [A,B] = complexCalculations(x,y); toc
Elapsed time is 126.810807 seconds.

Phew. Let's run it again:

>> tic; [A,B] = complexCalculations(x,y); toc
Elapsed time is 0.275681 seconds.


Welcome to cache.m

complexCalculations.m is internally cached, as so:

function [A,B] = complexCalculations(x,y)

% first we hash the inputs:
temp.x = x; temp.y = y;
hash = DataHash(temp);

% we ask cache if its seen this data before
cached_data = cache(hash);

if ~isempty(cached_data)
    % wow. cache.m has cached the results. let's just return that 
    A = cached_data.A;
    B = cached_data.B;

% do some complicated calculation.
% ... 
% ...

% cache the results:
cached_data = struct;
cached_data.A = A;
cached_data.B = B;

Any function can be internally cached in this manner. You can also use cache anything by hash:

A = hugeCalculations(B); % takes a super long time

and retrieve it easily with:

A = cache(DataHash(B)); % super fast

cache stores its cache locally in cached.mat.

If cache doesn't know what to translate a hash into, it will return an empty variable.

cache works best with cryptographic hashes, but you can use it with anything:


You can selectively clear entries from the hash table using


The best part? cache automatically manages its own size, and intelligently removes the least-used hashes first.


Get this repo

git clone