martinholters

travis_fold:start:worker_info
Worker information
hostname: 8178300b-eeeb-4ca7-9deb-de62eb27610b@1.production-1-worker-org-a-5-gce
version: v3.8.0 https://github.com/travis-ci/worker/tree/cccff7c89da7ba0cf47a90e5615266a45b372e30
instance: travis-job-4c088c18-8443-40fa-8e06-c9dc4353d13b travis-ci-garnet-trusty-1512502259-986baf0 (via amqp)
startup: 27.728081381s
travis_fold:end:worker_info
travis_fold:start:system_info
Build system information
Build language: cpp

martinholters

travis_fold:start:worker_info
Worker information
hostname: i-044ffdf-precise-production-2-worker-org-docker.travisci.net:e02d8c1d-2839-4dc7-8583-ca58236d1b7c
version: v2.5.0 https://github.com/travis-ci/worker/tree/da3a43228dffc0fcca5a46569ca786b22991979f
instance: 6c3bd47:travis:default
startup: 1.892870923s
travis_fold:end:worker_info
travis_fold:start:system_info
Build system information
Build language: cpp

martinholters

Worker information
hostname: i-020909f-precise-production-2-worker-org-docker.travisci.net:ecab4d7d-9896-485f-afbc-984bb6cce1af
version: v2.5.0 https://github.com/travis-ci/worker/tree/da3a43228dffc0fcca5a46569ca786b22991979f
instance: d3e685e:travis:default
startup: 393.097645ms
Build system information
Build language: cpp
Build group: stable
Build dist: precise
Build id: 200053652

martinholters

using BenchmarkTools

function dobench(n, nd)
    x = falses(ntuple(x -> n, nd)...)
    @benchmark slicedim($x, 1, 3)
end

function dobench()
    println(dobench(5, 1))
    println(dobench(10000, 1))

martinholters

travis_fold:start:worker_info
Worker information
hostname: ip-10-12-2-135:05c3e975-b454-400f-a84c-904c081cd1d4
version: v2.5.0-8-g19ea9c2 https://github.com/travis-ci/worker/tree/19ea9c20425c78100500c7cc935892b47024922c
instance: 645c69b:travis:default
startup: 1.99580549s
travis_fold:end:worker_info
travis_fold:start:system_info
Build system information
Build language: cpp

martinholters

This is the output of the Benchmark code for APL-style indexing Compat code.

The "time plain array" is the time needed to index into a plain array is nano seconds per indexing operation. The relative times are relative to this and either for indexing into a prefabricated ArrayAPL or indexing with @compat.

Julia Version 0.3.2
Platform Info:
  System: Linux (x86_64-linux-gnu)
 CPU: Intel(R) Xeon(R) CPU E5-1620 v2 @ 3.70GHz

martinholters

using Compat

function idxbench(N::Integer, idx...)
    a=rand(repmat([3], length(idx))...)
    idxbench(a, N, idx...)
end

function idxbench(a::AbstractArray, N::Integer, idx...)
    b=Compat.ArrayAPL(a)