Predicting the effects of parallelism on an algorithm is central in designing efficient applications. In this seminar, Avah presents the models used to determine the effects of parallelism and concurrency. She Continue reading
One Comment
Author: aserio
Phylanx Report: April 2019
The twentieth month of work focused on work in the following areas:
- Performance improvements
- Keras Backend
- hpxMP improvements
Phylanx Seminar: Distributed Object in HPX
Often when writing distributed applications, a user will want to represent a large data set as a single object, despite not fitting in memory on a single node. In order to address this use case, Weile Wei and Maxwell Reeser Continue reading
Phylanx Report: March 2019
The nineteenth month of work focused on work in the following areas:
- Traveler Team Visit (March 7th and 8th)
- Support for the Keras Back-end
Phylanx Seminar: Ordering
This seminar presents Avah’s work on high-dimensional random orders, a well known structure in Computational Geometry and useful for computing Skyline queries in multi-objective optimization. The talk works through definitions, theory and some examples. Talk materials can be found below: Continue reading
Phylanx Report: February 2019
The eighteenth month of work focused on work in the following areas:
- Keras backend support
- Primitive implementations
- 3D tensor support
- Documentation support
Phylanx Report: January 2019
The seventeenth month of work focused on work in the following areas:
- CMake Fixes
- Adding primitives to support Keras backend
- Implementing support for 3D tensors
- Adding Sphinx Documentation support
Phylanx Seminar: Parallelism and Distributed Arrays
In this seminar, Avah presents her latest model of the tiling problem in Phylanx. The lecture covers the assumptions made, constrains, and the open questions yet to be solved. Lecture materials can be found below: Continue reading
Phylanx Report: December 2018
The sixteenth month of work focused on work in the following areas:
- Added 3D tensor support to node_data
- Implementing primitives needed for Keras backend (Issue: #684)
- Modularize Traveler to better support multiple platforms