Digging into MLPerf Benchmark Suite to Inform AI Infrastructure Decisions

By John Russell

April 9, 2019

With machine learning and deep learning storming into the datacenter, the new challenge is optimizing infrastructure choices to support diverse ML and DL workflows. How many accelerators do you (really) need? Which interconnect scheme works best? How do various frameworks compare on different compute architectures? When can you choose less expensive CPUs and rely on the GPUs to do the magic? How different are the needs of training versus inferencing?

“There is no one server that does the job perfectly well,” said Ramesh Radhakrishnan, distinguished engineer, Dell EMC, to a packed session[I] at GTC last month. “You see a variety of servers used to execute these kinds of workloads.” Precisely to this point, a flurry of benchmarking tools is emerging to help make sense of ML/DL performance requirements and optimizations. There’s the Deep500 with grand ambitions but still very nascent and aimed mostly at very large scale systems. There are early movers – DeepBench, TF_CNN_Bench, and DAWNBench, for example – with typically narrower strengths and notable shortfalls. More recently, MLPerf has started emerging as popular tool that borrows from those coming before it.

MLPerf, a broader ML/DL benchmarking effort supported by industry and academia, is gaining a foothold at least for assessing training workloads. In his GTC session, Demystifying Deep Learning Infrastructure Choices Using MLPerf Benchmark Suite, Radhakrishnan took the audience on a test drive through MLPerf by presenting performance data from testing on four different systems (1-to-8 GPUs) with differing topologies He also offered comments about the other benchmarks and their contribution to progress so far.

Yes the GPUs profiled are all from Nvidia and the systems from Dell EMC – the occasion was GTC after all – but Radhakrishnan’s even-handed approach made for a solid primer. At HPCwire’s request, Nvidia agreed to make the link to Radhakrishnan’s streamed session public ahead of schedule and in time for inclusion in this article. It’s 44 minutes well-spent for those seeking a MLPerf overview. A few of the session highlights are covered here along with key slides from Radhakrishnan’s session.

Given the sudden rise of ML and DL it is perhaps not surprising that benchmarking tools have started sprouting. Each has value. “[Take] TF_CNN_Bench (TensorFlow convolutional neural network benchmark), for example,” said Radhakrishnan. “It’s very commonly used and people use it for different types of GPUs and it works well but is focused on a single domain, convolutional networks. If you are deploying a translation-based network that’s using RNN (recurrent neural networks), can you take the same observations that you made using a CNN (convolutional neural network) and assume the same thing for your other model? [No], there’s going to be differences. So you want to have a wide coverage on domains.”

DeepBench from Baidu does cover different domains. “It is primarily used for measuring the performance of the core operations that happen in your neural networks. [It has] CNN and RNN coverage [but] if you want to look at system level performance, this is not a tool for that. It doesn’t account for software frameworks, it doesn’t account for distributed training – you need the right set of tools for that,” noted Radhakrishnan.

Enter MLPerf. Early last year, “the folks that introduced DAWNBench from Stanford University worked with Google, Baidu, AMD, Intel and others to come up with MLPerf. This addressed a lot of shortcomings in previous benchmarks. It has coverage of different domains. In terms of metrics it borrows from DAWNBench, which uses execution time to [attain accuracy]. You are using Docker containers and producing all the information needed along with providing data sets so anyone can reproduce these [MLPerf] results,” said Radhakrishnan.

Importantly, emphasized Radhakrishnan, MLPerf has extensive support from industry and universities. At this writing more than 30 companies and roughly ten academic institutions are listed on the MLPerf website. This is key to culling out bias and driving MLPerf’s ability to make fair comparisons between systems or between two different ML/DL elements, according to Radhakrishnan. It also gives the organization “freedom to enable innovation to happen.”

So what is MLPerf? The organization describes it as, “A broad ML benchmark suite for measuring performance of ML software frameworks, ML hardware accelerators, and ML cloud platforms.” The current release, version 0.5, has five different domains: image classification, object detection, language translation, reinforcement learning, and recommendation. The MLPerf metric of choice – speedup – is the amount to time that it takes train a neural net model in these domains to state of the art target accuracy.

You can see the models used for each domain on the slide. Generally they represent state of the art models said Radhakrishnan – ResNet-50 used in image identification, is a deep (50 layer) residual neural network introduced in 2015. RNN GNMT (Google Neural Machine Translation, introduced in 2016) and Transformer (an NLP approach introduced in 2017) are likewise leading edge. One of MLPerf’s goals, said Radhakrishnan, is stay very current.

“It’s a very agile development model. There were actually seven or eight domains initially considered; these five made it [into the spec], but you can definitely see more domains being added and those models being refreshed based on what’s current and state of the art,” he said. “MLPerf is definitely a needed tool today and [we’re] really excited about looking to see how it changes over times and expands and adds domains as. It covers training now but I think they are really close to releasing the inference workload as well.”

The first MLPerf published results were in December. Nvidia, no surprise, trumpeted its strong performance. (See HPCwire article, Nvidia Leads Alpha MLPerf Benchmarking Round.) Radhakrishnan suggested those results didn’t have enough data points to examine some of the questions Dell EMC and colleagues at the University of Texas (College of Engineering) were interested in. His slide deck, which will be available soon, is good way to get a fast overview of the exercise.

“What we did in our work is actually run MLPerf against a wide range of platforms, GPU platforms, starting from a workstation with a single CPU and two GPUs, all the way to [a server with] eight GPUs. We decided to do a whole suite of tests ranging across scaling, comparing different GPU models, and looking at interconnect technologies whether it is PCIe or NVLink to help answer some of those questions,” he said.

Shown below is a kind of summary slide showing system performance based on number of GPUs used. The MLPerf score is how much faster the system executes compared to the MLPerf reference “which in this case is a single P100 instance and that P100 is not very well optimized either,” said Radhakrishnan. The smallest machine, based on Quadro GV100, scored 14.

“The score of 14 means this two-GPU platform ran 14 times faster than one P100.  However you should not make the assumption you are going to see that speed difference between P100s and GV100s every time. What this lets you do is you can actually calculate, because now that your baseline is the same, you can make the assumption that an eight-GPU platform performs four time more efficiently than my two-GPU platform. So you can make assumptions on scale. We are seeing some of the benchmarks scale linearly, so something like SSD (single shot multi-box detector) and other image classifications, get really good speed up going from two-to-four-to-eight GPUs. Some of the other benchmarks, recommendation and translation, don’t see that same efficiency and scale,” he said.

The tasks/features outlined in orange in the slide below are the areas benchmarked by Dell EMC.

“One question which always comes up is which framework should I use. Since there is so much information on Tensorflow from Google and Mxnet from Nvidia, we took those two submissions (from the initial MLPerf 0.20 run) and recreated those runs. We did this on a four-GPU NVLink platform, and plotted accuracy against time. Mxnet actually completed the target accuracy in about 245 min whereas TensorFlow took about 265 minutes. One of the reasons they are so close to each other is they take advantage of the Nvidia cuDNN library. In this case we actually looked at the GPU kernels that were being implemented and found that here were eight kernels called by both of these frameworks, they were exactly the same kernels, and they account for about 40 percent on average, of the execution time,” said Radhakrishnan.

For the purposes of his presentation, Radhakrishnan introduced two hypothetical characters Dave, a data scientist concerned only with performance, and Sally, a datacenter infrastructure manager concerned with cost, efficiency, and satisfying diverse constituencies. He discussed how their needs differ and how MLPerf can inform both.

“I showed Dave these results and he said “are you sure your Mxnet is optimized and running properly because I thought tensor flow was slower?” He was right in one way. Because this part of the submission that was done used TensorFlow version 1.12 [which] introduced XLA compiling. What XLA compiling does is basically does tensor fusion where it takes the graph and optimizes some of the optimized kernels that for your particular infrastructure, right, or your GPUs. So when we ran the same run without XLA compiler enabled, it actually took almost 200 minutes longer. One key takeaway is to make sure you are using the right optimizations, so no matter what you have underneath. If you aren’t taking advantage a compiling or using the right options you are going to leave a lot of performance on the table,” said Radhakrishnan.

One interesting result was the dramatic impact being able to use mixed precision during training. Radhakrishnan’s team looked Tesla V100-PCIe performance when used in precision and mixed precision modes.

“With Volta you get the capability of using mixed precision training. What this means is it uses a technique where it combines half-precision floating point and single-precision in order to improve performance and keep the same accuracy you get in your single precision frame. In addition to memory savings, you are also using dedicated hardware on V100s to do 4×4 matrix multiplies. However, there are some coding changes involved to take advantage of this,” noted Radhakrishnan.

As is shown in the slide below, it is worth the effort. Here improvements of roughly 150 percent to 330 percent were achieved. “A Resnet-50 job on an eight-GPU system that runs in about ten minutes if you take advantage of mixed precision; if you don’t it take advantage of it, [it takes] close to 500 minutes. After I prepared this slide yesterday, Jensen Huang (Nvidia CEO) announced you are getting this with practically just one or two lines change in your code. There’s an API exposed and once you enable that, it does different things you are supposed to do to make sure you still maintain the same accuracy,” said Radhakrishnan.

Radhakrishnan presented results from four different interconnect schemes including use of NVLink with GPUs using SXM2 form factor (permits high clock rate), use of PCIe switches, and PCIe standalone. Radhakrishnan noted the last arrangement (scheme on far right of slide) allows users to take a container “pin it to a CPU and then use a GPU” and that this provides the best CPU-to-GPU bandwidth and that’s useful for some applications though he didn’t specify examples.

“If you look at how machine learning performs across all of these designs. If you look at translation benchmarks you see that improves performance by anywhere from 40-to-50 percent and when you look at it in terms of minutes, that’s 20 minutes and 30 minutes. It is a sizeable improvement when you actually look at the percentage. There’s [also] improvement for MaskRCNN. When it comes image classification, you are seeing that the peer-to-peer capability provided by the PCIe switch makes a difference; you actually get difference between that platform and the two where you cannot do peer-to-peer.

“So system design does matter and as an infrastructure provider you definitely want to make sure to assign the tasks that are doing a lot of GPU-to-GPU communication, to the right platform. If you had a mix of platforms in your datacenter, if you have workloads that are actually doing a lot of peer-to-peer like translation benchmark, you want to make sure they get queued up on your NVLink platforms. For some of the other benchmarks, the differences that you get are not that significant,” said Radhakrishnan 

The bottom line, according to Radhakrishnan, is MLPerf provides a needed tool for assessing infrastructure choices as well as ML choices. “Use the tools to monitor and actually figure out how many GPUs you need [because] throwing as many GPUs as you can at the problem does not result in a linear increase in your performance. There are some workloads in which you can definitely see improvement and there are others where you are better off running at one or two GPUs,” he said.

The same is true for interconnect. “We looked at what NVLink brings to the table. For a two GPU system there are a couple of benchmarks for which NVLink did not make a difference,” he said but noted on larger systems NVLink can be critical when performance is your top priority although even there performance can vary by the model used.

There’s quite a bit more in Radhakrishnan’s presentation, including for example deeper dives into CPU utilization – it turns out TensorFlow is a CPU hog – and GPU utilization. It will be interesting to watch MLPerf’s development and how widely adopted it becomes.

Slide source: GTC2019

Link to Radhakrishnan’s GTC streamed (slides and audio) session: https://on-demand.gputechconf.com/gtc/2019/video/_/s9553-demystifying-deep-learning-infrastructure-choices-using-mlperf-benchmark-suite/

[i]Demystifying Deep Learning Infrastructure Choices Using MLPerf Benchmark Suite

Abstract: We’ll describe a new benchmark suite proposed by the Deep Learning community for machine learning workloads. We’ll present a quantitative analysis of an early version (0.5) of benchmark known as MLPerf and explain performance impact of NVIDIA GPU architecture across a range of DL applications. This work includes evaluating MLPerf performance on Turing, Volta, and Pascal to demonstrate the performance impact of NVIDIA GPU architecture across a range of DL applications. We’ll evaluate the impact of system-level technologies — Nvlink vs. PCIe topology — using server- and workstation-class platforms to show how system architecture impacts DL training workloads. We also plan to discuss our work to characterize MLPerf benchmark performance using profiling tools (GPU, CPU, memory & I/O), our hyperparameter tuning study (batch size, learning rate, SGD optimizer) on MLPerf performance, and map real world application use cases to MLPerf suite and how to quantify results for specific DL practitioner use cases. https://gputechconf2019.smarteventscloud.com/connect/sessionDetail.ww?SESSION_ID=263091

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!

RPI Powers Up ‘AiMOS’ AI Supercomputer

December 11, 2019

Designed to push the frontiers of computing chip and systems performance optimized for AI workloads, an 8 petaflops (Linpack) IBM Power9-based supercomputer has been unveiled in upstate New York that will be used by IBM Read more…

By Doug Black

At SC19: Developing a Digital Twin

December 11, 2019

In the not too distant future, we can expect to see our skies filled with unmanned aerial vehicles (UAVs) delivering packages, maybe even people, from location to location. In such a world, there will also be a digita Read more…

By Aaron Dubrow

Supercomputers Help Predict Carbon Dioxide Levels

December 10, 2019

The Earth’s terrestrial ecosystems – its lands, forests, jungles and so on – are crucial “sinks” for atmospheric carbon, holding nearly 30 percent of our annual CO2 emissions as they breathe in the carbon-rich Read more…

By Oliver Peckham

Finally! SC19 Competitors Live and in Color!

December 10, 2019

You know the saying “better late than never”? That’s how my cluster competition coverage is faring this year. With SC19 coming late in November, quickly followed by my annual trip to South Africa to cover their clu Read more…

By Dan Olds

Intel’s Jim Clarke on its New Cryo-controller and why Intel isn’t Late to the Quantum Party

December 9, 2019

Intel today introduced the ‘first-of-its-kind’ cryo-controller chip for quantum computing and previewed a cryo-prober tool for characterizing quantum processor chips. The new controller is a mixed-signal SoC named Ho Read more…

By John Russell

AWS Solution Channel

Making High Performance Computing Affordable and Accessible for Small and Medium Businesses with HPC on AWS

High performance computing (HPC) brings a powerful set of tools to a broad range of industries, helping to drive innovation and boost revenue in finance, genomics, oil and gas extraction, and other fields. Read more…

IBM Accelerated Insights

GPU Scheduling and Resource Accounting: The Key to an Efficient AI Data Center

[Connect with LSF users and learn new skills in the IBM Spectrum LSF User Community!]

GPUs are the new CPUs

GPUs have become a staple technology in modern HPC and AI data centers. Read more…

What’s New in HPC Research: Natural Gas, Precision Agriculture, Neural Networks and More

December 6, 2019

In this bimonthly feature, HPCwire highlights newly published research in the high-performance computing community and related domains. From parallel programming to exascale to quantum computing, the details are here. Read more…

By Oliver Peckham

RPI Powers Up ‘AiMOS’ AI Supercomputer

December 11, 2019

Designed to push the frontiers of computing chip and systems performance optimized for AI workloads, an 8 petaflops (Linpack) IBM Power9-based supercomputer has Read more…

By Doug Black

Intel’s Jim Clarke on its New Cryo-controller and why Intel isn’t Late to the Quantum Party

December 9, 2019

Intel today introduced the ‘first-of-its-kind’ cryo-controller chip for quantum computing and previewed a cryo-prober tool for characterizing quantum proces Read more…

By John Russell

On the Spack Track @SC19

December 5, 2019

At the annual supercomputing conference, SC19 in Denver, Colorado, there were Spack events each day of the conference. As a reflection of its grassroots heritage, nine sessions were planned by more than a dozen thought leaders from seven organizations, including three U.S. national Department of Energy (DOE) laboratories and Sylabs... Read more…

By Elizabeth Leake

Intel’s New Hyderabad Design Center Targets Exascale Era Technologies

December 3, 2019

Intel's Raja Koduri was in India this week to help launch a new 300,000 square foot design and engineering center in Hyderabad, which will focus on advanced com Read more…

By Tiffany Trader

AWS Debuts 7nm 2nd-Gen Graviton Arm Processor

December 3, 2019

The “x86 Big Bang,” in which market dominance of the venerable Intel CPU has exploded into fragments of processor options suited to varying workloads, has n Read more…

By Doug Black

Ride on the Wild Side – Squyres SC19 Mars Rovers Keynote

December 2, 2019

Reminding us of the deep and enabling connection between HPC and modern science is an important part of the SC Conference mission. And yes, HPC is a science its Read more…

By John Russell

NSCI Update – Adapting to a Changing Landscape

December 2, 2019

It was November of 2017 when we last visited the topic of the National Strategic Computing Initiative (NSCI). As you will recall, the NSCI was started with an Executive Order (E.O. No. 13702), that was issued by President Obama in July of 2015 and was followed by a Strategic Plan that was released in July of 2016. The question for November of 2017... Read more…

By Alex R. Larzelere

Tsinghua University Racks Up Its Ninth Student Cluster Championship Win at SC19

November 27, 2019

Tsinghua University has done it again. At SC19 last week, the eight-time gold medal-winner team took home the top prize in the 2019 Student Cluster Competition Read more…

By Oliver Peckham

Using AI to Solve One of the Most Prevailing Problems in CFD

October 17, 2019

How can artificial intelligence (AI) and high-performance computing (HPC) solve mesh generation, one of the most commonly referenced problems in computational engineering? A new study has set out to answer this question and create an industry-first AI-mesh application... Read more…

By James Sharpe

D-Wave’s Path to 5000 Qubits; Google’s Quantum Supremacy Claim

September 24, 2019

On the heels of IBM’s quantum news last week come two more quantum items. D-Wave Systems today announced the name of its forthcoming 5000-qubit system, Advantage (yes the name choice isn’t serendipity), at its user conference being held this week in Newport, RI. Read more…

By John Russell

DARPA Looks to Propel Parallelism

September 4, 2019

As Moore’s law runs out of steam, new programming approaches are being pursued with the goal of greater hardware performance with less coding. The Defense Advanced Projects Research Agency is launching a new programming effort aimed at leveraging the benefits of massive distributed parallelism with less sweat. Read more…

By George Leopold

Ayar Labs to Demo Photonics Chiplet in FPGA Package at Hot Chips

August 19, 2019

Silicon startup Ayar Labs continues to gain momentum with its DARPA-backed optical chiplet technology that puts advanced electronics and optics on the same chip Read more…

By Tiffany Trader

SC19: IBM Changes Its HPC-AI Game Plan

November 25, 2019

It’s probably fair to say IBM is known for big bets. Summit supercomputer – a big win. Red Hat acquisition – looking like a big win. OpenPOWER and Power processors – jury’s out? At SC19, long-time IBMer Dave Turek sketched out a different kind of bet for Big Blue – a small ball strategy, if you’ll forgive the baseball analogy... Read more…

By John Russell

Cray, Fujitsu Both Bringing Fujitsu A64FX-based Supercomputers to Market in 2020

November 12, 2019

The number of top-tier HPC systems makers has shrunk due to a steady march of M&A activity, but there is increased diversity and choice of processing compon Read more…

By Tiffany Trader

Crystal Ball Gazing: IBM’s Vision for the Future of Computing

October 14, 2019

Dario Gil, IBM’s relatively new director of research, painted a intriguing portrait of the future of computing along with a rough idea of how IBM thinks we’ Read more…

By John Russell

Intel Debuts New GPU – Ponte Vecchio – and Outlines Aspirations for oneAPI

November 17, 2019

Intel today revealed a few more details about its forthcoming Xe line of GPUs – the top SKU is named Ponte Vecchio and will be used in Aurora, the first plann Read more…

By John Russell

Leading Solution Providers

SC 2019 Virtual Booth Video Tour


Kubernetes, Containers and HPC

September 19, 2019

Software containers and Kubernetes are important tools for building, deploying, running and managing modern enterprise applications at scale and delivering enterprise software faster and more reliably to the end user — while using resources more efficiently and reducing costs. Read more…

By Daniel Gruber, Burak Yenier and Wolfgang Gentzsch, UberCloud

Dell Ramps Up HPC Testing of AMD Rome Processors

October 21, 2019

Dell Technologies is wading deeper into the AMD-based systems market with a growing evaluation program for the latest Epyc (Rome) microprocessors from AMD. In a Read more…

By John Russell

Cray Wins NNSA-Livermore ‘El Capitan’ Exascale Contract

August 13, 2019

Cray has won the bid to build the first exascale supercomputer for the National Nuclear Security Administration (NNSA) and Lawrence Livermore National Laborator Read more…

By Tiffany Trader

SC19: Welcome to Denver

November 17, 2019

A significant swath of the HPC community has come to Denver for SC19, which began today (Sunday) with a rich technical program. As is customary, the ribbon cutt Read more…

By Tiffany Trader

When Dense Matrix Representations Beat Sparse

September 9, 2019

In our world filled with unintended consequences, it turns out that saving memory space to help deal with GPU limitations, knowing it introduces performance pen Read more…

By James Reinders

With the Help of HPC, Astronomers Prepare to Deflect a Real Asteroid

September 26, 2019

For years, NASA has been running simulations of asteroid impacts to understand the risks (and likelihoods) of asteroids colliding with Earth. Now, NASA and the European Space Agency (ESA) are preparing for the next, crucial step in planetary defense against asteroid impacts: physically deflecting a real asteroid. Read more…

By Oliver Peckham

Cerebras to Supply DOE with Wafer-Scale AI Supercomputing Technology

September 17, 2019

Cerebras Systems, which debuted its wafer-scale AI silicon at Hot Chips last month, has entered into a multi-year partnership with Argonne National Laboratory and Lawrence Livermore National Laboratory as part of a larger collaboration with the U.S. Department of Energy... Read more…

By Tiffany Trader

Jensen Huang’s SC19 – Fast Cars, a Strong Arm, and Aiming for the Cloud(s)

November 20, 2019

We’ve come to expect Nvidia CEO Jensen Huang’s annual SC keynote to contain stunning graphics and lively bravado (with plenty of examples) in support of GPU Read more…

By John Russell

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