research
∙
05/20/2022
(Poly)Logarithmic Time Construction of Round-optimal n-Block Broadcast Schedules for Broadcast and irregular Allgather in MPI
We give a fast(er), communication-free, parallel construction of optimal...
share
research
∙
09/26/2021
A Doubly-pipelined, Dual-root Reduction-to-all Algorithm and Implementation
We discuss a simple, binary tree-based algorithm for the collective allr...
research
∙
10/29/2020
A more Pragmatic Implementation of the Lock-free, Ordered, Linked List
The lock-free, ordered, linked list is an important, standard example of...
research
∙
08/27/2020
k-ported vs. k-lane Broadcast, Scatter, and Alltoall Algorithms
In k-ported message-passing systems, a processor can simultaneously rece...
research
∙
05/19/2020
Efficient Process-to-Node Mapping Algorithms for Stencil Computations
Good process-to-compute-node mappings can be decisive for well performin...
research
∙
01/20/2020
High-Quality Hierarchical Process Mapping
Partitioning graphs into blocks of roughly equal size such that few edge...
research
∙
10/29/2019
Decomposing Collectives for Exploiting Multi-lane Communication
Many modern, high-performance systems increase the cumulated node-bandwi...
research
∙
07/30/2019
How to Make the Preconditioned Conjugate Gradient Method Resilient Against Multiple Node Failures
We study algorithmic approaches for recovering from the failure of sever...
research
∙
03/28/2019
More Parallelism in Dijkstra's Single-Source Shortest Path Algorithm
Dijkstra's algorithm for the Single-Source Shortest Path (SSSP) problem ...
research
∙
05/21/2018
Stamp-it: A more Thread-efficient, Concurrent Memory Reclamation Scheme in the C++ Memory Model
We present Stamp-it, a new, concurrent, lock-less memory reclamation sch...
research
∙
04/20/2018
Parallel Quicksort without Pairwise Element Exchange
Standard implementations of 2-way, parallel, distributed memory Quicksor...
research
∙
03/12/2018
Memory Models for C/C++ Programmers
The memory model is the crux of the concurrency semantics of shared-memo...
research
∙
12/17/2017
A new and five older Concurrent Memory Reclamation Schemes in Comparison (Stamp-it)
Memory management is a critical component in almost all shared-memory, c...
research
∙
11/23/2017
