Smarter Worksharing

Monday, February 12th, 2018

An article titled “Smarter Worksharing” that highlights our research at Linnaeus University appeared in HiPEAC INFO 53 (page 31). This issue of HiPEAC INFO was dedicated to machine learning and AI research in Europe. Sabri Pllana, Linnaeus University, explains how his team is using machine learning for optimal worksharing on heterogeneous computing systems.

HiPEAC is the European Network on High Performance and Embedded Architecture and Compilation. HiPEAC INFO is quarterly published news by HiPEAC.

Further reading suggestions:

Suejb Memeti and Sabri Pllana. ‘A machine learning approach for accelerating DNA sequence analysis’. International Journal of High Performance Computing Applications, published online on June 26, 2016


MemAxes: Visualization and Analytics for Characterizing Complex Memory Performance Behaviors

Wednesday, August 30th, 2017

A new publication in the journal: IEEE Transactions on Visualization and Computer Graphics by our senior lecturer Ilir Jusufi about High-Performance Computing memory performance analysis through the use of Visual Analytics.

Abstract: Memory performance is often a major bottleneck for high-performance computing (HPC) applications. Deepening memory hierarchies, complex memory management, and non-uniform access times have made memory performance behavior difficult to characterize, and users require novel, sophisticated tools to analyze and optimize this aspect of their codes. Existing tools target only specific factors of memory performance, such as hardware layout, allocations, or access instructions. However, today’s tools do not suffice to characterize the complex relationships between these factors. Further, they require advanced expertise to be used effectively. We present MemAxes, a tool based on a novel approach for analytic-driven visualization of memory performance data. MemAxes uniquely allows users to analyze the different aspects related to memory performance by providing multiple visual contexts for a centralized dataset. We define mappings of sampled memory access data to new and existing visual metaphors, each of which enabling a user to perform different analysis tasks. We present methods to guide user interaction by scoring subsets of the data based on known performance problems. This scoring is used to provide visual cues and automatically extract clusters of interest. We designed MemAxes in collaboration with experts in HPC and demonstrate its effectiveness in case studies.

Reference: A. Gimenez; T. Gamblin; I. Jusufi; A. Bhatele; M. Schulz; P. T. Bremer; B. Hamann, “MemAxes: Visualization and Analytics for Characterizing Complex Memory Performance Behaviors,” in IEEE Transactions on Visualization and Computer Graphics.

Download the full publication here.

//Diana Unander Nordle