aprof new minor release 0.4.1

Changes in the new release

Imports testthat for automated testing. Minor bugs and spelling changes

Installation

Release on GITHUB. CRAN release to follow very soon.

aprof new minor release 0.3.9

Changes in the new release

Now using automated testing, and some code changes in favor of robustness.

Installation

Release on GITHUB. CRAN release to follow very soon.

aprof new minor release 0.3.2

Changes in the new release

  1. Fixed the bug which was kindly reported by Mark Miller. In short, a regex did not recognize the correct number of repeating strings of digits (in the memory profiling output) . This caused the identification of memory statistics from the profiler output to fail. Now fixed by explicitly stating the repeating pattern.

Installation

Release on GITHUB. CRAN release to follow very soon.

BES workshop on code efficiency

Dear R-coders,

We are happy to organise the workshop “Speeding up ecological and evolutionary computations in R; essentials of high performance computing for biologists.” The workshop is to be held over the lunch break on Tuesday, 15 December.

On behalf my co-organisers, I would like to introduce the workshop and provide some details. The course will start with a general overview in the form of a 45-minute lecture, we will then proceed into break-out sessions geared towards more individual preferences and needs (20-35 minutes).

Due to the short time slot made available for workshops over the lunch break, the workshop is limited to providing an overview of available techniques towards high-performance computing, and showing you how to identify bottlenecks in your programs. However, in the break-out session we also aim to provide a chance to apply the theory presented in the lecture section, or at least ask targeted questions.

We cannot tailor the break-out session to suite each person’s individual needs, and therefore we will break out into three groups. The potential groups will include I) profiling and identifying bottlenecks in code, II) parallel computing and III) extending R with C. To make best use of the time, we require everyone to download the tutorial (S1_text) from the following publication:

Visser et al. Plos Comp 2015

Further specific requirements include,

Under I), please make sure you bring a laptop with R and the package aprof installed.

Under II), we will focus on multi-cored solutions (e.g. parallelcomputing on your laptop) not on distributed computing (e. i. a cluster or network of workstations) for practical reasons. You obviously will need to bring a multi-cored laptop.

Under III), some experience with C is needed. You will also need to work through section 5 of S1_Text for your OS.

We have been informed that conference organisers will not be able to supply extension cords, so make sure you have a fully charged battery. We are looking forward to an active and engaging meeting with you!

Your Sincerely,

Marco Visser, Caspar Hallmann, Eelke Jongejans and Sean McMahon.

Note: The latest version of aprof can be downloaded here or from CRAN.

aprof new major release 0.3.1

Changes in the new release

  1. Support for memory profiling, including basic summary statistics using the default print semantics.
  2. Standard aprof ploting (plot.aprof) now recognizes memory profiled output and add this to the standard line plot.

Installation

Release on GITHUB. CRAN release to follow very soon.