We’re excited to announce that the newest version of our Bitfusion Ubuntu 14 Tensorflow AMI is now available in the AWS Marketplace with Tensorflow 1.0! This Tensorflow release marks a major milestone of functionality, robustness, and -- going forward -- backwards compatibility.
We’re excited to announce the newest AMI in our lineup: Bitfusion Ubuntu 14 MXNet AMI, now available in the AWS Marketplace. Amazon CTO Werner Vogels has recently announced that MXNet is the company's official deep learning framework of choice, paving the way for fast growth and interest this previously lesser known project.
We’re excited to announce that the newest version of our Bitfusion Ubuntu 14 Tensorflow AMI is now available in the AWS Marketplace. We’ve also made some exciting new changes to our pricing model to help save you money.
Join us November 15-17 in Salt Lake City, Utah as we partake in the 2016 Supercomputing Conference. In conjunction with introducing the latest version of our revolutionary Bitfusion Boost software, which now comes with a free trial for up to 4 node deployments, we will be presenting a talk titled: GPU Virtualization: Optimizing Performance and Cost of your Applications, on Wednesday November 16 at 11am, Nimbix Booth #4072.
TensorFlow GPU performance on AWS p2 instances is between 2x-3x faster when compared to previous generation g2 instances across a variety of convolutional neural networks. Recently, we made our Bitfusion Deep Learning AMIs available on the newly announced AWS P2 instances. Naturally, one of the first questions that arises is, how does the performance of the new P2 instances compare to that of the the previous generation G2 instances. In this post we take a quick look at single-GPU performance across a variety of convolutional neural networks. To keep things consistent we start each EC2 instance with the exact same AMI, thus keeping the driver, cuda, cudnn, and framework the same across the instances. TensorFlow GPU Performance To evaluate TensorFlow performance we utilized the Bitfusion TensorFlow AMI along with the convnet-benchmark to measure for forward and backward propagation times for some of the more well known convolutional neural networks including AlexNet, Overfeat, VGG, and GoogleNet. Because of the much larger GPU memory of 12 GiB, the P2 instances can accommodate much larger batch sizes than the G2 instances. For the purpose of the benchmarks below, the batch sizes were selected for each network type such that they could run on the G2 as well as on the P2 instances. The Tables below summarize the results obtained for G2 and P2 instances: Bitfusion Ubuntu 14 TensorFlow AMI Launch on AWS! g2.2xlarge - Nvidia K520 Network Batch Size Forward Pass (ms) Backward Pass (ms) Total Time (ms) AlexNet 512 502 914 1416 Overfeat 256 1134 2934 4068 VGG 64 750 2550 3300 GoogleNet 128 600 1587 2187 p2.xlarge - Nvidia K80 Network Batch Size Forward Pass (ms) Backward Pass (ms) Total Time (ms) AlexNet 512 254 462 716 Overfeat 256 427 847 1274 VGG 64 423 869 1292 GoogleNet 128 341 783 1124 Averaging the speedup across all four types of networks, the results show an approximate ~2.42x improvement in performance - not bad for an instance which is only ~1.39 more expensive on an hourly on demand basis. We have several other Deep learning AMIs available in the AWS Marketplace including Caffe, Chainer, Theano, Torch, and Digits. If you are interested in seeing GPU Performance benchmarks for any of the above drop us a note. Are you currently developing AI applications, but spending too much time wrangling machines and setting up your infrastructure? We are currently offering a Free 30-Day Trial of Bitfusion Flex!
Leads and prospects generated from search or display ads can be very costly and challenging to obtain. That's why Apposphere, an Austin, Texas based company, saw a huge opportunity in mining the huge amount of information and activity happening on social media for lead generation.