HPC Technique Propels Deep Learning at Scale

By Tiffany Trader

February 21, 2017

Researchers from Baidu’s Silicon Valley AI Lab (SVAIL) have adapted a well-known HPC communication technique to boost the speed and scale of their neural network training and now they are sharing their implementation with the larger deep learning community.

The technique, a modified version of the OpenMPI algorithm “ring all-reduce,” is being used at Baidu to parallelize the training of their speech recognition model, Deep Speech 2, across many GPU nodes. The two pieces of software Baidu is announcing today are the baidu-allreduce C library, as well as a patch for TensorFlow, which allows people who have already modeled in TensorFlow to compile this new version and use it for parallelizing across many devices. The codes are available on GitHub.

Ring all-reduce – all GPUs send data simultaneously

Baidu’s SVAIL team developed the approach about two years ago for their internal deep learning framework, named Gene and Majel (in tribute to the famous Star Trek creator and the actress who voiced the onboard computer interfaces for the series). The technique is commonplace in HPC circles, but underused within artificial intelligence and deep learning, according to Baidu.

Many of the researchers in the SVAIL group had come from the high performance computing space and recognized the competitive edge it offered.

“The algorithm is actually part of OpenMPI, but the OpenMPI implementation is not as fast,” comments Baidu Research Scientist Shubho Sengupta. “So the way we stumbled upon it was we started using OpenMPI for doing training and we realized it was not scaling to the extent that we want it to scale. I started digging through the OpenMPI source, found the algorithm, saw that it’s not very efficient, and reimplemented it.”

The SVAIL researchers wrote their own implementation of the ring algorithm for higher performance and better stability. The key distinction from the OpenMPI version is that the SVAIL implementation avoids extraneous copies between the CPU and GPU.

Explains Sengupta, “Once OpenMPI does the communication of these matrices, if the matrices are in GPU memory, it actually copies to CPU memory to do the reduction part of it – that’s actually quite wasteful. You don’t really need to do a copy, you could just write a small kernel that does the reduction in GPU memory space itself. And this especially helps when you are doing all-reduce within a node and all the GPUs are within a PCI root complex, then it doesn’t do any of the copies actually – it can just do everything in GPU memory space. This very simple idea of eliminating this copy resulted in this speedup in scaling over OpenMPI’s own implementation.”

Employing this algorithm along with SVAIL’s focus on fast networking (InfiniBand) and careful hardware-software codesign has enabled the team to get linear GPU scaling up to 128 GPUs, an achievement that was detailed in their December 2015 paper, “Deep Speech 2: End-to-End Speech Recognition in English and Mandarin.”

With their internal implementation of ring all-reduce, the team achieves between a 2.3-21.4X speedup over OpenMPI (version 1.8.5) depending on the number of GPUs.

Sengupta notes that their implementation is fastest for a small number of GPUs. “At 8 GPUs it’s about 20x faster, then as you increase the number of GPUs, it drops because now you actually have to copy data to the CPU to send across the network. But for the internal framework, we can scale all the way up to 128 GPUs and get linear scaling.”

Comparison of two different all-reduce implementations. All times are in seconds. Performance gain is the ratio of OpenMPI all-reduce time to SVAIL’s all-reduce time. (Source: Deep Speech 2 paper)

Sengupta’s teammate Baidu Research Scientist Andrew Gibiansky says similar benefits can now be seen with TensorFlow: “In terms of the TensorFlow implementation, we get the same linear scaling path past eight. In terms of a comparison with running on a single GPU, it ends up being about 31x faster at 40 GPUs.”

After the Deep Speech 2 paper was published, the SVAIL team began getting requests from the community who wanted to know more about the implementation. Given that the algorithm is pretty tightly coupled to SVAIL’s proprietary deep learning framework, they needed to come up with a different way to release it, so they created two new implementations, one specifically for TensorFlow and one that is more general.

Gibiansky, who led the work on the TensorFlow patch, describes their multi-pronged approach to disseminating the information. “You can read the blog post [for a thorough technical explanation] and figure it out. If you’re using TensorFlow, you can use our modification to train your own models with this. And if you’re a deep learning author, you can look at our C library and integrate that. The goal is really to take this idea we’ve found to be really successful internally and try to start spreading it so that other people can also take advantage of it.”

Sengupta shares an interesting perspective on the opportunities to be mined for deep learning within HPC.

“With MPI – people [in deep learning] think that it is this old technology, that it is not relevant, but I think because of our work we have shown that you can build very fast collectives using MPI and that allows you to do synchronous gradient descent which converges faster, gives you deterministic results and you don’t need to do asynchronous gradient descent with parameter servers which was the dominant way of doing this when we first started,” says Sengupta.

As for the reduced-copy approach propagating back to MPI, Gibiansky notes that if you look at some of the other MPI implementations, they’re slowly moving their collectives to GPU versions. “MVPICH recently introduced an all-gather that doesn’t end up copying to CPU – so OpenMPI will probably get there, it just might take a while. Potentially giving this a little more visibility, we can spur that on.”

“There’s a lot of interest now in collectives and one thing we also realized is the all-reduce operation used in traditional HPC setups, it actually transfers data that’s actually not very large,” Sengupta adds. “What it usually does, when I talk to HPC people, it’s trying to figure out the status of something across a bunch of machines – while in deep learning we are transferring these large matrices – like 2048×2048, essentially 4 million 32-bit floating points. For the traditional HPC community, this is a very atypical input for all-reduce. The traditional HPC community does not actually use all-reduce with really large data sizes. I think with deep learning, more and more people are realizing that collective operations for really large matrices is also very important.”

A detailed explanation of ring all-reduce and Baidu’s GPU implementation is covered in this technical blog post, published today by Baidu Research. A variant of the technique is also used to provide high-performance node-local scaling for PaddlePaddle, the company’s open source deep learning framework.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

San Diego Supercomputer Center to Welcome ‘Expanse’ Supercomputer in 2020

July 18, 2019

With a $10 million dollar award from the National Science Foundation, San Diego Supercomputer Center (SDSC) at the University of California San Diego is procuring a new supercomputer, called Expanse, to be deployed next Read more…

By Staff report

Informing Designs of Safer, More Efficient Aircraft with Exascale Computing

July 18, 2019

During the process of designing an aircraft, aeronautical engineers must perform predictive simulations to understand how airflow around the plane impacts flight characteristics. However, modeling the complexities and su Read more…

By Rob Johnson

How Fast is Your Rubik Solver; This One’s Probably Faster

July 18, 2019

In the race to solve Rubik’s Cube, the time-to-finish keeps shrinking. This year Philipp Weyer from Germany won the 10th World Cube Association (WCA) Championship held in Melbourne, Australia, with a 6.74-second perfo Read more…

By John Russell

HPE Extreme Performance Solutions

Bring the Combined Power of HPC and AI to Your Business Transformation

A growing number of commercial businesses are implementing HPC solutions to derive actionable business insights, to run higher performance applications and to gain a competitive advantage. Read more…

IBM Accelerated Insights

Smarter Technology Revs Up Red Bull Racing

In 21st century business, companies that effectively leverage their information resources – thrive. As it turns out, the same is true in Formula One racing. Read more…

Intel Debuts Pohoiki Beach, Its 8M Neuron Neuromorphic Development System

July 17, 2019

Neuromorphic computing has received less fanfare of late than quantum computing whose mystery has captured public attention and which seems to have generated more efforts (academic, government, and commercial) but whose Read more…

By John Russell

Informing Designs of Safer, More Efficient Aircraft with Exascale Computing

July 18, 2019

During the process of designing an aircraft, aeronautical engineers must perform predictive simulations to understand how airflow around the plane impacts fligh Read more…

By Rob Johnson

Intel Debuts Pohoiki Beach, Its 8M Neuron Neuromorphic Development System

July 17, 2019

Neuromorphic computing has received less fanfare of late than quantum computing whose mystery has captured public attention and which seems to have generated mo Read more…

By John Russell

Goonhilly Unveils New Immersion-Cooled Platform, Doubles Down on Sustainability Mission

July 16, 2019

Goonhilly Earth Station has opened its new datacenter – an enhancement to its existing tier 3 facility – in Cornwall, England, touting an ambitious commitme Read more…

By Oliver Peckham

ISC19 Cluster Competition: Application Results, Finally!

July 15, 2019

Our exhaustive coverage of the ISC19 Student Cluster Competition continues as we discuss the application scores below. While the scores were typically high, som Read more…

By Dan Olds

Nvidia Expands DGX-Ready AI Program to 19 Countries

July 11, 2019

Nvidia’s DGX-Ready Data Center Program, announced in January and designed to provide colo and public cloud-like options to access the company’s GPU-powered Read more…

By Doug Black

Argonne Team Makes Record Globus File Transfer

July 10, 2019

A team of scientists at Argonne National Laboratory has broken a data transfer record by moving a staggering 2.9 petabytes of data for a research project.  The data – from three large cosmological simulations – was generated and stored on the Summit supercomputer at the Oak Ridge Leadership Computing Facility (OLCF)... Read more…

By Oliver Peckham

Nvidia, Google Tie in Second MLPerf Training ‘At-Scale’ Round

July 10, 2019

Results for the second round of the AI benchmarking suite known as MLPerf were published today with Google Cloud and Nvidia each picking up three wins in the at Read more…

By Tiffany Trader

Applied Materials Embedding New Memory Technologies in Chips

July 9, 2019

Applied Materials, the $17 billion Santa Clara-based materials engineering company for the semiconductor industry, today announced manufacturing systems enablin Read more…

By Doug Black

High Performance (Potato) Chips

May 5, 2006

In this article, we focus on how Procter & Gamble is using high performance computing to create some common, everyday supermarket products. Tom Lange, a 27-year veteran of the company, tells us how P&G models products, processes and production systems for the betterment of consumer package goods. Read more…

By Michael Feldman

Cray, AMD to Extend DOE’s Exascale Frontier

May 7, 2019

Cray and AMD are coming back to Oak Ridge National Laboratory to partner on the world’s largest and most expensive supercomputer. The Department of Energy’s Read more…

By Tiffany Trader

Graphene Surprises Again, This Time for Quantum Computing

May 8, 2019

Graphene is fascinating stuff with promise for use in a seeming endless number of applications. This month researchers from the University of Vienna and Institu Read more…

By John Russell

AMD Verifies Its Largest 7nm Chip Design in Ten Hours

June 5, 2019

AMD announced last week that its engineers had successfully executed the first physical verification of its largest 7nm chip design – in just ten hours. The AMD Radeon Instinct Vega20 – which boasts 13.2 billion transistors – was tested using a TSMC-certified Calibre nmDRC software platform from Mentor. Read more…

By Oliver Peckham

TSMC and Samsung Moving to 5nm; Whither Moore’s Law?

June 12, 2019

With reports that Taiwan Semiconductor Manufacturing Co. (TMSC) and Samsung are moving quickly to 5nm manufacturing, it’s a good time to again ponder whither goes the venerable Moore’s law. Shrinking feature size has of course been the primary hallmark of achieving Moore’s law... Read more…

By John Russell

Deep Learning Competitors Stalk Nvidia

May 14, 2019

There is no shortage of processing architectures emerging to accelerate deep learning workloads, with two more options emerging this week to challenge GPU leader Nvidia. First, Intel researchers claimed a new deep learning record for image classification on the ResNet-50 convolutional neural network. Separately, Israeli AI chip startup Hailo.ai... Read more…

By George Leopold

Nvidia Embraces Arm, Declares Intent to Accelerate All CPU Architectures

June 17, 2019

As the Top500 list was being announced at ISC in Frankfurt today with an upgraded petascale Arm supercomputer in the top third of the list, Nvidia announced its Read more…

By Tiffany Trader

Top500 Purely Petaflops; US Maintains Performance Lead

June 17, 2019

With the kick-off of the International Supercomputing Conference (ISC) in Frankfurt this morning, the 53rd Top500 list made its debut, and this one's for petafl Read more…

By Tiffany Trader

Leading Solution Providers

ISC 2019 Virtual Booth Video Tour

CRAY
CRAY
DDN
DDN
DELL EMC
DELL EMC
GOOGLE
GOOGLE
ONE STOP SYSTEMS
ONE STOP SYSTEMS
PANASAS
PANASAS
VERNE GLOBAL
VERNE GLOBAL

Intel Launches Cascade Lake Xeons with Up to 56 Cores

April 2, 2019

At Intel's Data-Centric Innovation Day in San Francisco (April 2), the company unveiled its second-generation Xeon Scalable (Cascade Lake) family and debuted it Read more…

By Tiffany Trader

Cray – and the Cray Brand – to Be Positioned at Tip of HPE’s HPC Spear

May 22, 2019

More so than with most acquisitions of this kind, HPE’s purchase of Cray for $1.3 billion, announced last week, seems to have elements of that overused, often Read more…

By Doug Black and Tiffany Trader

A Behind-the-Scenes Look at the Hardware That Powered the Black Hole Image

June 24, 2019

Two months ago, the first-ever image of a black hole took the internet by storm. A team of scientists took years to produce and verify the striking image – an Read more…

By Oliver Peckham

Announcing four new HPC capabilities in Google Cloud Platform

April 15, 2019

When you’re running compute-bound or memory-bound applications for high performance computing or large, data-dependent machine learning training workloads on Read more…

By Wyatt Gorman, HPC Specialist, Google Cloud; Brad Calder, VP of Engineering, Google Cloud; Bart Sano, VP of Platforms, Google Cloud

Why Nvidia Bought Mellanox: ‘Future Datacenters Will Be…Like High Performance Computers’

March 14, 2019

“Future datacenters of all kinds will be built like high performance computers,” said Nvidia CEO Jensen Huang during a phone briefing on Monday after Nvidia revealed scooping up the high performance networking company Mellanox for $6.9 billion. Read more…

By Tiffany Trader

Chinese Company Sugon Placed on US ‘Entity List’ After Strong Showing at International Supercomputing Conference

June 26, 2019

After more than a decade of advancing its supercomputing prowess, operating the world’s most powerful supercomputer from June 2013 to June 2018, China is keep Read more…

By Tiffany Trader

In Wake of Nvidia-Mellanox: Xilinx to Acquire Solarflare

April 25, 2019

With echoes of Nvidia’s recent acquisition of Mellanox, FPGA maker Xilinx has announced a definitive agreement to acquire Solarflare Communications, provider Read more…

By Doug Black

It’s Official: Aurora on Track to Be First US Exascale Computer in 2021

March 18, 2019

The U.S. Department of Energy along with Intel and Cray confirmed today that an Intel/Cray supercomputer, "Aurora," capable of sustained performance of one exaf Read more…

By Tiffany Trader

  • arrow
  • Click Here for More Headlines
  • arrow
Do NOT follow this link or you will be banned from the site!
Share This