Skip to content

Latest commit

 

History

History
49 lines (34 loc) · 2.02 KB

RELEASE.md

File metadata and controls

49 lines (34 loc) · 2.02 KB

Release 1.0.0

Features

  • Removed Chameleon dependency completely.

Release 0.1.4

Features

  • LU factorization (hicma_zgetrf) on double complex matrices stored in tile low-rank (TLR) format. Link to branch.

Release 0.1.3

Features

  • Support for a new application that simulates the 3D unstructured mesh deformation of a population of the novel coronaviruses (i.e., SARS-CoV-2). This relies on the virus geometry extracted from the Protein Data Bank (PDB) codenamed PDBID 6VXX available at (https://www.rcsb.org/structure/6VXX). The corresponding basis function and data geometry are available at STARS-H (https://github.com/ecrc/stars-h).

Release 0.1.2

Features

  • New matrix kernel (3D exponential) for statistics application from STARS-H (https://github.com/ecrc/stars-h) is supported.
  • Sequential operations on low-rank tiles are moved into a standalone library called HCORE (https://github.com/ecrc/hcore).
  • Number of flops is reported. See the driver routines under timing folder.
  • Multiplication order of tiles in each inner product performed by HiCMA_GEMM can be changed via adding --reorderinnerproducts flag while calling timing/time_zgemm_tile. This flag enables sorting tiles according to increasing sum of ranks of tile pairs that which will be multiplied.
  • More tests in timing/CMakeLists.txt

Bug Fixes and Other Changes

  • Set minimum number of singular vectors in HCORE_GEMM to 1 instead of 2

Release 0.1.1

Features

  • Triangular solve (left) for low rank A and B/X
  • Triangular solve (left, lower) for low rank A and full rank B/X
  • Testing routine for POSV and TRSM
  • HCORE routine for C=C+AB where A is low rank and B and C are full rank
  • Separate routines for dense and low rank matrix generation

Release 0.1.0

Features

  • Matrix-Matrix Multiplication
  • Cholesky Factorization
  • Double Precision
  • Task-based Programming Models
  • Shared and Distributed-Memory Environments
  • Support for StarPU Dynamic Runtime Systems
  • Testing Suite and Examples