#### Filter Results:

- Full text PDF available (22)

#### Publication Year

2004

2017

- This year (3)
- Last 5 years (17)
- Last 10 years (31)

#### Publication Type

#### Co-author

#### Journals and Conferences

Learn More

- James Demmel, Laura Grigori, Mark Hoemmen, Julien Langou
- SIAM J. Scientific Computing
- 2012

We present parallel and sequential dense QR factorization algorithms that are both optimal (up to polylogarithmic factors) in the amount of communication they perform and just as stable as… (More)

- Mark Hoemmen
- 2010

1

Current iterative methods for solving linear equations assume reliability of data (no “bit flips”) and arithmetic (correct up to rounding error). If faults occur, the solver usually either aborts, or… (More)

- Marghoob Mohiyuddin, Mark Hoemmen, James Demmel, Katherine A. Yelick
- Proceedings of the Conference on High Performance…
- 2009

Data communication within the memory system of a single processor node and between multiple nodes in a system is the bottleneck in many iterative sparse matrix solvers like CG and GMRES. Here… (More)

- James Demmel, Mark Hoemmen, Marghoob Mohiyuddin, Katherine A. Yelick
- 2008 IEEE International Symposium on Parallel and…
- 2008

The performance of sparse iterative solvers is typically limited by sparse matrix-vector multiplication, which is itself limited by memory system and network performance. As the gap between… (More)

- James Demmel, Laura Grigori, Mark Hoemmen, Julien Langou
- ArXiv
- 2008

We present parallel and sequential dense QR factorization algorithms that are optimized to avoid communication. Some of these are novel, and some extend earlier work. Communication includes both… (More)

- James Demmel, Mark Hoemmen, +5 authors Katherine A. Yelick
- 2007

Our goal is to minimize the communication costs of Krylov Subspace Methods (KSMs) to solve either Ax = b or Ax = λx, when A is a large sparse matrix. By communication costs we mean both bandwidth and… (More)

interface. The Failable interface has methods for marking, unmarking, and checking whether the object’s data are allowed to experience bit flips. FTGMRES mark failability of the relevant objects on… (More)

- Ichitaro Yamazaki, Sivasankaran Rajamanickam, Erik G. Boman, Mark Hoemmen, Michael A. Heroux, Stanimire Tomov
- SC14: International Conference for High…
- 2014

Krylov subspace projection methods are widely used iterative methods for solving large-scale linear systems of equations. Researchers have demonstrated that communication-avoiding (CA) techniques can… (More)

- Eric Bavier, Mark Hoemmen, Sivasankaran Rajamanickam, Heidi Thornquist
- Scientific Programming
- 2012

Solvers for large sparse linear systems come in two categories: direct and iterative. Amesos2, a package in the Trilinos software project, provides direct methods, and Belos, another Trilinos… (More)