GPU Optimization
GPU Optimization Topics • Threading Details – Wavefronts and warps – Thread scheduling for both AMD and NVIDIA GPUs – Predication • Optimziation – Thread mapping – Device occupancy – Vectorization 2 Work Groups to HW Threads • OpenCL kernels are structured into work groups that map to device compute units • Compute units on GPUs consist of SIMT processing elements • Work groups automatically get b
GPU Accelerators Radically Boost SQL Queries
June 26, 2015 Timothy Prickett Morgan In-memory databases are all the rage for very fast query processing, but you have to have the right balance of compute and memory for queries against in-memory databases to really scream. Sometimes, a large number of relatively small cores and a reasonably beefy chunk of RAM can radically outperform clusters with hundreds or thousands of server nodes running
Introduction to Neural Machine Translation with GPUs (part 1) | NVIDIA Technical Blog
Introduction to Neural Machine Translation with GPUs (part 1) Note: This is the first part of a detailed three-part series on machine translation with neural networks by Kyunghyun Cho. You may enjoy part 2 and part 3. Neural machine translation is a recently proposed framework for machine translation based purely on neural networks. This post is the first of a series in which I will explain a simp
Chainer – A flexible framework of neural networks — Chainer 7.8.1 documentation
Chainer – A flexible framework of neural networks¶ Chainer is a powerful, flexible and intuitive deep learning framework. Chainer supports CUDA computation. It only requires a few lines of code to leverage a GPU. It also runs on multiple GPUs with little effort. Chainer supports various network architectures including feed-forward nets, convnets, recurrent nets and recursive nets. It also supports
GPU-Accelerated Graph Analytics in Python with Numba | NVIDIA Technical Blog
Numba is an open-source just-in-time (JIT) Python compiler that generates native machine code for X86 CPU and CUDA GPU from annotated Python Code. (Mark Harris introduced Numba in the post Numba: High-Performance Python with CUDA Acceleration.) Numba specializes in Python code that makes heavy use of NumPy arrays and loops. In addition to JIT compiling NumPy array code for the CPU or GPU, Numba ex
TheanoをWindowsにインストール - 人工知能に関する断創録
Deep Learningを実装するのによく使われるTheanoというPythonライブラリをWindowsマシンにインストールしたのでそのときの記録。ただ使うだけだったらPythonとnumpy/scipyをインストールした後にpip installs Theanoで普通に使えていた。実際、GPUが貧弱なMacbook Airではこの方法でインストールしていた。今回、PCを買い替えた（2015/1/19）こともあって、NVIDIA社のGPUを使って高速演算できるようにしてみたというわけ。私のマシンは、 OS : Windows 8.1 64bit GPU: NVIDIA GeForce GTX 760 Ti OEM という環境。Theano TutorialにもWindowsへのインストール方法は載っているけどはっきり言ってよくわからない。いろいろ調べていたところ以下の記事が自信満々で手
Installing CUDA, OpenCL, and PyOpenCL on AWS EC2
This post describes how to setup CUDA, OpenCL, and PyOpenCL on EC2 with Ubuntu 12.04. It took me a while to get working properly, so what follows are the steps I took to install CUDA and OpenCL on EC2. These steps should work for any machine with a CUDA capable card. If you'd just like to create an instance with everything preinstalled, I've created a public AMI named Ubuntu-12.04-GPU-CUDA-OpenCL-
Deep Learning on Amazon EC2 GPU with Python and nolearn - PyImageSearch
Last week I wrote a post detailing my experience with CUDAMat, Deep Belief Networks, and Python using my MacBook Pro. The post is fairly long and full of screenshots to document my experience. But the gist of it is this: Even after installing the NVIDIA Cuda SDK and configuring CUDAMat, my CPU was training my Deep Belief Network (implemented by nolearn) faster than my GPU. As you can imagine, I wa
Deep Dynamic Neural Networks for Gesture Segmentation and Recognition
The purpose of this paper is to describe a novel method called Deep Dynamic Neural Networks(DDNN) for the Track 3 of the Chalearn Looking at People 2014 challenge [1]. A generalised semi-supervised hierarchical dynamic framework is proposed for simultaneous gesture segmentation and recognition taking both skeleton and depth images as input modules. First, Deep Belief Networks(DBN) and 3D Convoluti
NVIDIA cuDNN | NVIDIA Developer
Accelerated Learning cuDNN provides kernels, targeting Tensor Cores, to deliver best available performance on compute-bound operations. It offers heuristics for choosing the right kernel for a given problem size. Expressive Op Graph API The user defines computations as a graph of operations on tensors. The cuDNN library has both a direct C API and an open-source C++ frontend for convenience. Most
AWSでGPUを試したら単精度1TFLOPSでた - ロードバランスすだちくん
EC2にもGPUが搭載されているインスタンスがあるけど、科学技術計算用途に使えるのか試してみました。 結論から書くと、Marketplaceに用意されているNVIDIAのAMIは古くてなんとなく嫌なので、全部セットアップしてサンプル流してって軽く考えていたら大変だった。。。 結果的にはGPU1ノードで単精度1TFLOPSは出せたのでなかなか良いのではないのでしょうか？ まずEC2に用意されているGPUインスタンスは2種類しかありません。今回はg2.2xlargeを利用します。 Marketplaceからお手軽に試したい人は以下のリンクからどうぞ。 https://aws.amazon.com/marketplace/search/results/?page=1&searchTerms=GPU いい感じのAMIが無い美しいCUDA環境を作りたいが為にCentOS6.5を使いたいわけですが、先
GPU-Accelerated Computing with Python
GPU-Accelerated Computing with Python NVIDIA’s CUDA Python provides a driver and runtime API for existing toolkits and libraries to simplify GPU-based accelerated processing. Python is one of the most popular programming languages for science, engineering, data analytics, and deep learning applications. However, as an interpreted language, it’s been considered too slow for high-performance computi
PyCUDAの紹介
11. スペックIntel Xeon X5570 × 264bit 4コアNVIDIA Tesla M2050 × 2448GPUコア単精度ピーク時1.03Tflops倍精度対応Memory 22GB高速なネットワーク同一Placement Group内は高速通信可能CentOS 5.5同程度のマシンを買うと80〜100万円ぐらい？ 12. 費用On-Demand Instance$2.1 / hour (￥178.5 / hour)1年￥1,563,660，3年￥4,690,980，…Reserved Instance1年 $5,630 (￥478,550)$0.65 / hour (￥55 / hour)3年 $8,650 (￥735,250)$0.33 / hour (￥28 / hour)+EBS料金など(月$2〜)※$1 = ￥85で計算
PyMC3の計算でGPUを使っている気がしただけの話
たまには浮気させてください。PyMC3は内部でTheanoを使っており、自動微分（auto-diff）が計算可能でStanのサンプラーであるNUTSも実装済みです。またTheanoがGPUに対応しているため、これはMCMCの超高速化が簡単にできるのではッ！と試した記事になります。 まずは環境設定から。Windows 7 64bitにVisual Studio Express 2012, CUDA 6, Anaconda 2.0.0 (python 2.7 + numpyとか), theano 0.6.0, PyMC3をインストールします。手順が少し多い場合は分かる範囲でDAG構造を描くのがオススメです。学術書の章案内もDAG構造があるといいですね。さて、今回の手順の依存性をDAG構造で表すと以下になります。 1. Visual Studio Express 2012のインストール ここから
リリース、障害情報などのサービスのお知らせ
最新の人気エントリーの配信
j次のブックマーク
k前のブックマーク
lあとで読む
eコメント一覧を開く
oページを開く
http://www.issp.u-tokyo.ac.jp/public/keisan09/abstract.pdf
RGB-D Object Tracking:
A Particle Filter Approach on GPU
http://nbviewer.ipython.org/github/memelabo/JetsonMemo/blob/master/Jetson%20TK1%20IPython%20Notebook%20PyCUDA.ipynb
Netflix TechBlog
NVBIO: NVBIO
Parallel Implementation of Particle MCMC Methods on a GPU