MatsuLab. Lecture Note
ハイパフォーマンスコンピューティング †
- 日時
- 月曜日 １０：４５〜１２：１５（３〜４限）
- 場所
- 西８号館 ８３２号室
- 連絡
| 松岡教授 (Prof. S.Matsuoka) | matsu あっと is. |
| TA 金刺 (H.Kanezashi) | kanezashi あっと matsulab.is. |
目次 †
休講予定日 Lecture Cancelled †
10/19, 11/16, 02/08（補講はありません We do not have supplementary lectures）
授業概要と参考資料 Guidance and References †
発表スケジュール Schedule †
暫定的な割り当ては以下の通りですが、都合が悪い場合はTAまで希望日をメールしてください。
禁止リスト Inhibited List †
- Training Large Scale Deep Neural Networks on the Intel Xeon Phi Many-Core Coprocessor
- Memory fast-forward: A low cost special function unit to enhance energy efficiency in GPU for big data processing
- Optimized Deep Learning Architectures with Fast Matrix Operation Kernels on Parallel Platform
- Real-time anomaly detection in hyperspectral images using multivariate normal mixture models and GPU processing
- Hyperspectral Unmixing on GPUs and Multi-Core Processors: A Comparison
- Performance versus energy consumption of hyperspectral unmixing algorithms on multi-core platforms
- Optimizing communication and cooling costs in HPC data centers via intelligent job allocation
- Cost Minimization for Big Data Processing in Geo-Distributed Data Centers
- On Characterization of Performance and Energy Efficiency in Heterogeneous HPC Cloud Data Centers
- DaDianNao?: A Machine-Learning Supercomputer
- Mariana: tencent deep learning platform and its applications
- Performance Modeling and Scalability Optimization of Distributed Deep Learning Systems
- Asynchronous parallel stochastic gradient descent: a numeric core for scalable distributed machine learning algorithms
- FireCaffe?: near-linear acceleration of deep neural network training on compute clusters
- CA-SVM: Communication-Avoiding Support Vector Machines on Distributed Systems
- Large Scale Distributed Deep Networks
- Optimizing FPGA-based Accelerator Design for Deep Convolutional Neural Networks
- 24.77 Pflops on a Gravitational Tree-Code to Simulate the Milky Way Galaxy with 18600 GPUs
- Massively Parallel Models of the Human Circulatory System
- Moving to memoryland: in-memory computation for existing applications
- Intelligent SSD: A Turbo for Big Data Mining
- Scalable Multi-Access Flash Store for Big Data Analytics
- An FPGA-Based Tightly Coupled Accelerator for Data-Intensive Applications
- A reconfigurable fabric for accelerating large-scale datacenter services
- An FPGA-based In-Line Accelerator for Memcached
- Scaling Up the Training of Deep CNNs for Human Action Recognition
期末レポート Report †
- 期限 Due date: 02/08
- Summarize the general topic covering and including ALL THREE PAPERS regarding the state of the art in HPC and Big Data convergence.
- It should be 10 pages in IEEE conference paper format
- Please submit it to TA by email (NOT mailing list)
リンク Links †
![[PukiWiki]](image/pukiwiki.png "[PukiWiki]")
2015年度ハイパフォーマンスコンピューティング授業内容.pdf
