#### Filter Results:

- Full text PDF available (16)

#### Publication Year

2000

2018

- This year (1)
- Last 5 years (7)
- Last 10 years (8)

#### Publication Type

#### Co-author

#### Journals and Conferences

Learn More

- Paulius Micikevicius
- GPGPU
- 2009

In this paper we describe a GPU parallelization of the 3D finite difference computation using CUDA. Data access redundancy is used as the metric to determine the optimal implementation for both theâ€¦ (More)

A novel algorithm for encoding finite labeled trees is proposed in this paper. The algorithm establishes a one-to-one mapping between trees of order n and (n â€“ 2)-tuples of the node labels. Thisâ€¦ (More)

- I. Carpenter, Rick Archibald, +6 authors Mark A. Taylor
- IJHPCA
- 2013

The suitability of a spectral element based dynamical core (HOMME) within the Community Atmospheric Model (CAM) for GPU-based architectures is examined and initial performance results are reported.â€¦ (More)

In 1918 PrÃ¼fer showed a one-to-one correspondence between n-node labeled trees and (n â€“ 2)-tuples of node labels. The proof employed a tree code, computed by iteratively deleting the leaf with theâ€¦ (More)

- Paulius Micikevicius
- PDPTA
- 2004

Programmability and IEEE-standard floating point arithmetic makes the latest commodity graphics processors (GPUs) an attractive platform for general parallel computing. In this paper we describe theâ€¦ (More)

- Christopher B. Stapleton, Charles E. Hughes, J. Michael Moshell, Paulius Micikevicius, Marty Altman
- IEEE Computer
- 2002

T he lack of compelling content has relegated many promising entertainment technologies to laboratory curiosities. Although mixed-reality techniques show great potential, the entertainment businessâ€¦ (More)

- Paulius Micikevicius, Sharan Narang, +8 authors Hao Wu
- ArXiv
- 2017

Increasing the size of a neural network typically improves accuracy but also increases the memory and compute requirements for training the model. We introduce methodology for training deep neuralâ€¦ (More)

A queue-based PrÃ¼fer-like code is used to determine the expected number of level-i nodes in a random labeled tree on n nodes. Level-1 nodes are the leaves of a given tree and level-i nodes are leavesâ€¦ (More)

- Oreste Villa, Daniel R. Johnson, +9 authors William J. Dally
- SC14: International Conference for Highâ€¦
- 2014

Modern scientific discovery is driven by an insatiable demand for computing performance. The HPC community is targeting development of supercomputers able to sustain 1 ExaFlops by the year 2020 andâ€¦ (More)

In this paper we present O(n)-time algorithms for encoding/decoding n-node labeled trees as sequences of nâˆ’2 node labels. All known encodings of this type are covered, including PrÃ¼fer-like codes andâ€¦ (More)