IBM Research Achieves News Milestone in Deep Learning Performance

2017-08-08 17:21 Posted by: Jennifer in News Go to Comment

IBM just unveiled a new tool which is supposed to largely reduce the time spent in training distributed deep learning (DDL) system by using a lot of quality hardware to the task.

To be specific, this new technique tackles a major challenge when deploying deep learning, which is that a large neural network as well as large datasets can make deep learning thrive, but they also cause longer training time. It may be days or even weeks taken to train a large-scale AI model based on deep learning.

The scaled-up numbers of GPUs communicate with each other so this process usually spends a long term. As a matter of fact, when GPUs get faster, the process can be worse. It is indeed that faster GPUs learn faster, while via conventional software, the communications with one another cannot keep up. 

According to IBM’s Hillery Hunter, generally, the smarter as well as faster GPUs just need a better communication means, or these learners (GPUs) are out of sync and take most time to wait for each other’s results. Therefore, it means users may not get high speed, but potentially have degraded performance when using more, faster learners.

IBM’s new tech disclosed today can improve that problem right away. Even better, the new DDL software may run those popular open source codes such as Caffe, Tensorflow, Chainer, and Torch over a lot of neural networks and massive datasets with high accuracy and performance.  

IBM Research showed how to achieve the record communication overhead as well as 95% scaling efficiency on Caffe-based deep learning framework over 256 GPUs in the 64 IBM Power systems. Facebook AI Research set the previous scaling record that was near 90% efficiency for a Caffe2-based training, with higher communication overhead. 

In addition, IBM Research uses new DDL software to get new 33.8% image recognition accuracy for a neural network which was trained on a large-scale dataset including 7.5 million images, and the high image recognition accuracy was achieved in 7 hours only! 

How to achieve that? Hunter explained this, most deep learning frameworks normally scale up to multiple GPUs in one single server, rather than to multiple servers with certain GPUs. As for IBM Research, its team developed software as well as algorithms which can automate as well as optimize this complex and large-scale computing task over massive GPU accelerators which are attacked to a lot of servers. By utilizing the great power of these servers with massive GPUs, IBM Research made the milestone. 

There is no doubt that the new DDL software improves deep learning performance a lot. The company hopes users can see the advances in various AI applications, and is making DDL’s technical preview. DDL software now is available in its deep learning software distribution package, PowerAI 4. 

Tagged with:

Related Articles

ChinaNetCenter Net Profit Margin Decreases By 33% As Cloud Computing Vendors Jumping Into the Fray
10/16
China’s leading CDN platform provider ChinaNetCenter issued its performance report for the first quarters of the year last Friday, which shows a net income of 522 million to 614 million yuan, decreasing by 33 percent year-over-year to 43 percent.
Week 2017/10/09 Top 5 Web Hosting Industry News
10/13
we carefully collected the major events happening this week in the industry. Hope you can find something interesting things here.
Western Digital Makes New Breakthrough in HDD Technology
10/12
Western Digital develops a new magnetic recording technology with microwave assist.
Sina Weibo Suffers Outage Again, Caused by LUHAN’s Confession Tweet This Time
10/09
Chinese microblogging service Sina Weibo was briefly unavailable yesterday, reflecting sluggish internet speeds, the comment function was suspended and some pages cannot be displayed. The site appeared to be fully restored in less than two hours.
Weekly 2017/09/25 Top 5 Web Hosting Industry News
09/29
There were many exciting news and events happening in the industry this week, and here we conclude the top 5 for you.
Comment
Be the first to comment!
About the author
Jennifer
Views: 42635
Articles: 193

Author's Hot Articles

godaddy
hub

Hot Posts

Service Recommendation

Cloud Hosting
Cloud hosting is the fastest growing hosting solutions in the industry. »
Cloud Storage
Best Cloud Storage are named based on customer and expert reviews. »
SSL
The most popular SSL certificate providers in the industry are here. »
China Web Hosting
Find out a comprehensive evaluation to the host service provider in China. »
bluehost
Chat Advertising cooperation TOP