Posts
-
Phylanx Report: March 2018
The seventh month of work focused on work in the following areas:
Continue Reading -
Phylanx Seminar: Revealing the Magic of Blaze
Over the past several weeks, many team members have been asking how aspects of the Phylanx project are implemented. In this seminar, Hartmut explains the techniques used by Blaze to optimize matrix operations. He implements a matrix addition example which uses types, templates, and curiously recurring template patterns (CRTP) to reduce the number of temporaries made during the execution of the code. By avoiding these extra allocations in a compiler friendly way, Blaze can drastically reduce the amount of time it takes perform matrix operations. You can find links to the seminar materials below:
Continue Reading -
Phylanx Seminar: Polyhedral Optimizations
The polyhedral model is a mathematics framework used to optimize loop execution. Its use is supported by OpenScop, a specification for storing gathered information of an analyzed code. In this lecture, Rod discusses the potential use cases for the polyhedral model in Phylanx and the current state of the OpenScop file generator. You can find links to the seminar materials below:
Continue Reading -
Phylanx Report: February 2018
The sixth month of work focused on work in the following areas:
Continue Reading -
Phylanx Seminar: PhyFun
On February 23rd, Steve presented a Phylanx seminar on the current state of the Python front-end to Phylanx, PhyFun. This method of binding Phylanx primitives utilizes Python decorators. The talk describes how to use the decorators and some of the current implementation’s limitations. You can find the lecture as well as links to the seminar materials below:
Continue Reading -
Phylanx Seminar: PhySL
On February 9th, Hartmut presented another Phylanx seminar on the current state of the project. In addition to laying out the project in more detail, he described the special language PhySL which he and Parsa have been working on.
Continue Reading -
Introduction to Phylanx Coding
In this post I’ll go through the simple implementation of LRA (Logistic Regression Algorithm) to outline the Phylanx architecture and also demonstrate how one might go about writing their own programs in Phylanx. The complete version of the code discussed in this post can be found in the project’s GitHub repository under examples/algorithms directory and the corresponding dataset can be found here.
Continue Reading -
Phylanx Report: January 2018
The fifth month of work focused on work in the following areas:
Continue Reading -
Join the Team!
The Phylanx team is excited to announce a search for a postdoctoral researcher to help us to take our project’s development to the next level! We are now entering the sixth month of work and we find ourselves in need of more hands to continue our explosive rate of development. We are looking for a team member with strong C++ skills and experience and an interest in high performance computing, runtime systems, or machine learning. Familiarity with the paradigms and constructs of modern C++ is preferred. You can read more about the qualifications and apply for the position on LSU’s Careers Portal.
Continue Reading -
Phylanx Report: December 2017
The fourth month of work focused on work in the following areas:
Continue Reading -
Phylanx Seminar: The Agave Platform
Agave is a science-as-a-service platform which we are considering leveraging to assist the dissemination of our work. To help us understand what it can offer our team member, Steve, gave a presentation on the technology on December 11th.
Continue Reading -
Phylanx Report: November 2017
The third month of work focused on
Continue Reading -
Phylanx Seminar: Python Integration
Now that an HPX back-end is taking shape, the Phylanx team is starting to strategize how best to interface with a forthcoming python front-end. In a talk on November 3rd, Rod discussed the methods of interacting with Python in other projects, the current design of Phylanx, and a path to start building a Python interface.
Continue Reading -
Phylanx Report: October 2017
The second month of work focused on
Continue Reading -
Phylanx Seminar
Friday the Phylanx team at LSU held a seminar on the current theory, techniques, and methodology used by the Phylanx project. During the lecture Hartmut laid out the scope of the challenge we are trying to solve and the three components of the project as we have them today. Additionally, he explained the grammar we are using to describe expression trees and the current and future role Python is playing in the project.
Continue Reading -
Phylanx Report: September 2017
The first month of work was dedicated to the design and development of the first parts of the infrastructure needed for the overall project.
Continue Reading -
Project Kickoff Meeting
On August 24th and 25th the Phylanx team held our kickoff meeting at Louisiana State University. We had a productive meeting where we were able to start pinning down project details. Some of the major decisions include: formalizing the three tiered project structure, focusing on developing primitives, and drawing up a project timeline. You can find the notes that we took at the meeting below. In addition, we captured an image of our white board which shows the result of our timeline brainstorming
Continue Reading -
Hello World!
The Phylanx team is excited to start communicating our vision, research, and results of this project to the broader community. We believe that our ideas have the ability to influence the way that domain scientist utilize HPC resources. It is essential, therefore, to spread our message and include them in our work as early as possible . We hope that this platform will help us accomplish this goal. We plan to post regular updates about the project as well as elaborate on certain features of the code. In this way, we can complement the technical documentation provided in our GitHub repository with a more verbose and narrative driven format. Please feel free to comment and do not hesitate to reach out to us with questions or other inquiries.
Continue Reading
subscribe via RSS