Scaling graph-neural-network training with CPU-GPU clusters

In tests, new approach is 15 to 18 times as fast as predecessors.

Graphs are a useful way to represent data, since they capture connections between data items, and graph neural networks (GNNs) are an increasingly popular way to work on graphs. Common industry applications of GNNs include recommendation, search, and fraud detection.

The graphs used in industry applications are usually massive, with billions of nodes and hundreds of billions or even trillions of edges. Training GNNs on graphs of this scale requires massive memory storage and computational power, with correspondingly long training times and large energy footprints.

Related content
Information extraction, drug discovery, and software analysis are just a few applications of this versatile tool.

In a paper we’re presenting at this year’s KDD, my colleagues and I describe a new approach to distributed training of GNNs that uses both CPUs and GPUs, optimizing the allocation of tasks to different processor types to minimize training times.

In tests, our approach — DistDGLv2 — offered an 18-fold speedup over Euler, another distributed GNN training framework, on the same hardware. DistDGLv2 also achieves a speedup of up to 15-fold over distributed CPU training in a cluster of the same size.

Graph neural networks

In the GNN setting, graph nodes typically represent objects, and the graph edges represent relationships between objects. Both nodes and edges may have associated features — data such as object properties or types of relationships between objects.

Related content
Amazon’s George Karypis will give a keynote address on graph neural networks, a field in which “there is some fundamental theoretical stuff that we still need to understand.”

For each node in a graph, a GNN produces a vector representation (an embedding) that encodes information about the node and its neighborhood — often its one- or two-hop neighborhood, but sometimes larger regions. With the large graphs common in industrial applications, it can be time consuming to factor in all of a node’s one-hop neighbors, let alone its more distant neighbors. So when producing node embeddings, GNNs will often use minibatches of nodes sampled from the target node’s neighborhood.

There are many research works on minibatch sampling — for example, our global-neighbor-sampling technique, presented at KDD 2021. In our new paper, we implement a popular minibatch-sampling algorithm proposed by GraphSage, shown in the figure below. It first samples the target nodes (such as the blue node) and then samples their neighbor nodes (such as the red nodes and orange nodes). DistDGLv2, however, has the flexibility to implement other sampling algorithms.

Minibatch sampling procedure.png
An example of the minibatch sampling procedure.

DistDGLv2

DistDGLv2 has three main components:

  • a distributed key-value database (KVStore) to store node/edge features and learnable embeddings;
  • a distributed graph store to keep the partitioned graphs for minibatch sampling; and
  • a set of trainers to run forward and backward computation on minibatches to estimate the gradients of the model parameters.

To optimize the use of computational resources and scale to very large graphs, we divide these components between CPUs and GPUs. The distributed KVStore and graph store use CPU memory, and CPUs generate the minibatches. The trainers read the minibatch data into GPUs for minibatch computations.

Method overview.png
Method overview.

The key to accelerating minibatch training in DistDGLv2 is efficiently moving minibatches from CPU to GPU. To do this, DistDGLv2 deploys three strategies:

Related content
New method enables two- to 14-fold speedups over best-performing predecessors.

  • First, it uses the METIS graph-partitioning program (codeveloped by Amazon senior principal scientist George Karypis) to generate graph partitions with minimal edge cuts, and it collocates data with computation to reduce network communication;
  • It builds an asynchronous minibatch training pipeline to overlap computation and data movement in all hardware;
  • It moves as many computations to GPU as possible to take advantage of GPUs’ computational power.

To collocate data with computation, DistDGLv2 runs KVStore servers, distributed graph store servers, and trainers on the same set of machines. When a graph partition is loaded, its node and edge features go to the KVStore, and the graph structure goes to the graph store server. Each trainer is assigned a training set, where most training nodes and edges belong to the graph partition assigned to the same machine. In this way, most of the data associated with a minibatch will come from the local machine during the minibatch training.

Related content
Novel cross-graph-attention and self-attention mechanisms enable state-of-the-art performance.

DistDGLv2 implements the second and the third strategies by splitting the minibatch pipeline into seven stages, five of which help prepare a minibatch. We keep as many stages as possible on GPU to take advantage of GPUs’ computational power, while placing the minibatch sampling stages in CPU in another thread. This allows us to overlap minibatch computation in GPU and minibatch sampling in CPU.

As illustrated in the figure below, we run the last four stages in GPU; some of those stages are still involved in minibatch preparation.

In addition to this, we further overlap network communication and CPU computation. We have the sampling pipeline “look ahead” and sample multiple minibatches simultaneously. Thus, when a minibatch is being generated, while a given CPU is waiting for remote neighbor sampling (from another machine) or feature copy (to a GPU), it can move to another minibatch to sample neighbors or copying data locally. In this way, we can effectively hide network communication latency.

GNN training pipeline_.jpeg
The minibatch training pipeline, with a blowup of the minibatch generation step.

With these optimizations, DistDGLv2 can effectively perform distributed GNN training in a cluster of CPUs and GPUs. We demonstrate the efficiency of DistDGLv2 on a cluster of g4dn.metal instances with various GNN workloads. DistDGLv2’s performance relative to CPU-only methods indicates that GPUs can be more effective for distributed GNN minibatch training on massive graphs than CPUs.

Performance graph.png
A comparison of minibatch and full-graph training on the same hardware.

Researchers have also proposed using full graph training for GNN models. This method runs forward and backward computation on the entire graph. We did a comparison between minibatch training and full-graph training on the same graph datasets with the same hardware. We show that minibatch training is much more efficient to train GNN models, and the speed gap gets larger the larger the graphs grow.

On a graph built from the OGBN-papers100M dataset, which has 100 million nodes, minibatch training is about 100 times as fast. After six day’s training, full-graph training still cannot reach the same accuracy as minibatch training, while minibatch training takes 1.5 hours to reach the state-of-the-art performance on the same CPU.

Related content

US, WA, Seattle
We are a team of doers working passionately to apply cutting-edge advances in deep learning in the life sciences to solve real-world problems. As a Senior Applied Science Manager you will participate in developing exciting products for customers. Our team rewards curiosity while maintaining a laser-focus in bringing products to market. Competitive candidates are responsive, flexible, and able to succeed within an open, collaborative, entrepreneurial, startup-like environment. At the leading edge of both academic and applied research in this product area, you have the opportunity to work together with a diverse and talented team of scientists, engineers, and product managers and collaborate with others teams. Location is in Seattle, US Embrace Diversity Here at Amazon, we embrace our differences. We are committed to furthering our culture of inclusion. We have ten employee-led affinity groups, reaching 40,000 employees in over 190 chapters globally. We have innovative benefit offerings, and host annual and ongoing learning experiences, including our Conversations on Race and Ethnicity (CORE) and AmazeCon (gender diversity) conferences. Amazon’s culture of inclusion is reinforced within our 14 Leadership Principles, which remind team members to seek diverse perspectives, learn and be curious, and earn trust Balance Work and Life Our team puts a high value on work-life balance. It isn’t about how many hours you spend at home or at work; it’s about the flow you establish that brings energy to both parts of your life. We believe striking the right balance between your personal and professional life is critical to life-long happiness and fulfillment. We offer flexibility in working hours and encourage you to find your own balance between your work and personal lives Mentor & Grow Careers Our team is dedicated to supporting new members. We have a broad mix of experience levels and tenures, and we’re building an environment that celebrates knowledge sharing and mentorship. Our senior members enjoy one-on-one mentoring and thorough, but kind, code reviews. We care about your career growth and strive to assign projects based on what will help each team member develop into a better-rounded engineer and enable them to take on more complex tasks in the future. Key job responsibilities • Manage high performing engineering and science teams • Hire and develop top-performing engineers, scientists, and other managers • Develop and execute on project plans and delivery commitments • Work with business, data science, software engineer, biological, and product leaders to help define product requirements and with managers, scientists, and engineers to execute on them • Build and maintain world-class customer experience and operational excellence for your deliverables
US, Virtual
The Amazon Economics Team is hiring Interns in Economics. We are looking for detail-oriented, organized, and responsible individuals who are eager to learn how to work with large and complicated data sets. Some knowledge of econometrics, as well as basic familiarity with Stata, R, or Python is necessary. Experience with SQL, UNIX, Sawtooth, and Spark would be a plus. These are full-time positions at 40 hours per week, with compensation being awarded on an hourly basis. You will learn how to build data sets and perform applied econometric analysis at Internet speed collaborating with economists, data scientists and MBAʼs. These skills will translate well into writing applied chapters in your dissertation and provide you with work experience that may help you with placement. Roughly 85% of interns from previous cohorts have converted to full time economics employment at Amazon. If you are interested, please send your CV to our mailing list at econ-internship@amazon.com.
US, WA, Seattle
Amazon internships are full-time (40 hours/week) for 12 consecutive weeks with start dates in May - July 2023. Our internship program provides hands-on learning and building experiences for students who are interested in a career in hardware engineering. This role will be based in Seattle, and candidates must be willing to work in-person. Corporate Projects (CPT) is a team that sits within the broader Corporate Development organization at Amazon. We seek to bring net-new, strategic projects to life by working together with customers and evolving projects from ZERO-to-ONE. To do so, we deploy our resources towards proofs-of-concept (POCs) and pilot programs and develop them from high-level ideas (the ZERO) to tangible short-term results that provide validating signal and a path to scale (the ONE). We work with our customers to develop and create net-new opportunities by relentlessly scouring all of Amazon and finding new and innovative ways to strengthen and/or accelerate the Amazon Flywheel. CPT seeks an Applied Science intern to work with a diverse, cross-functional team to build new, innovative customer experiences. Within CPT, you will apply both traditional and novel scientific approaches to solve and scale problems and solutions. We are a team where science meets application. A successful candidate will be a self-starter comfortable with ambiguity, strong attention to detail, and the ability to work in a fast-paced, ever-changing environment. As an Applied Science Intern, you will own the design and development of end-to-end systems. You’ll have the opportunity to create technical roadmaps, and drive production level projects that will support Amazon Science. You will work closely with Amazon scientists, and other science interns to develop solutions and deploy them into production. The ideal scientist must have the ability to work with diverse groups of people and cross-functional teams to solve complex business problems.
US, WA, Seattle
Amazon internships are full-time (40 hours/week) for 12 consecutive weeks with start dates in May - July 2023. Our internship program provides hands-on learning and building experiences for students who are interested in a career in hardware engineering. This role will be based in Seattle, and candidates must be willing to work in-person. Corporate Projects (CPT) is a team that sits within the broader Corporate Development organization at Amazon. We seek to bring net-new, strategic projects to life by working together with customers and evolving projects from ZERO-to-ONE. To do so, we deploy our resources towards proofs-of-concept (POCs) and pilot programs and develop them from high-level ideas (the ZERO) to tangible short-term results that provide validating signal and a path to scale (the ONE). We work with our customers to develop and create net-new opportunities by relentlessly scouring all of Amazon and finding new and innovative ways to strengthen and/or accelerate the Amazon Flywheel. CPT seeks an Applied Science intern to work with a diverse, cross-functional team to build new, innovative customer experiences. Within CPT, you will apply both traditional and novel scientific approaches to solve and scale problems and solutions. We are a team where science meets application. A successful candidate will be a self-starter comfortable with ambiguity, strong attention to detail, and the ability to work in a fast-paced, ever-changing environment. As an Applied Science Intern, you will own the design and development of end-to-end systems. You’ll have the opportunity to create technical roadmaps, and drive production level projects that will support Amazon Science. You will work closely with Amazon scientists, and other science interns to develop solutions and deploy them into production. The ideal scientist must have the ability to work with diverse groups of people and cross-functional teams to solve complex business problems.
US, CA, Palo Alto
The Amazon Search team creates powerful, customer-focused search solutions and technologies. Whenever a customer visits an Amazon site worldwide and types in a query or browses through product categories, Amazon Search services go to work. We design, develop, and deploy high performance, fault-tolerant distributed search systems used by millions of Amazon customers every day. We’re seeking a Principal Scientist with a deep expertise in Search Science. Your responsibilities will include everything from developing and prototyping innovative machine learning, and deep learning algorithms to implementing, testing, and supporting full solutions in a production environment. We are looking for innovators who can contribute to advancing search technology on what’s scientifically possible while remaining committed to creating world-class products. Joining this team, you’ll experience the benefits of working in a dynamic, entrepreneurial environment, while leveraging the resources of Amazon.com (AMZN), Earth's most customer-centric company one of the world's leading internet companies. We provide a highly customer-centric, team-oriented environment in our offices located in Palo Alto, California. Key job responsibilities As a hands-on leader of this team, you’ll be responsible for defining key research questions, identifying relevant data, adopting or proposing innovative machine learning solutions conducting rigorous experiments, publishing results and working with the engineering team to deploy these solutions. As a strategic leader, you will identify investment opportunities, develop long term strategies, and propose, prioritize and deliver on goals. You’ll also participate in organizational planning, hiring, mentorship and leadership development. You will be technically fearless and with a passion for building scalable science and engineering solutions. You will serve as a key scientific resource in full-cycle development (conception, design, implementation, testing to documentation, delivery, and maintenance). About the team Starting in 2009, the Visual Search & Augmented Reality team has thus far launched many visual search solutions on the Amazon App that use computer vision and machine learning/deep learning to help customers complete their shopping missions more easily; multiple internal teams at Amazon (devices, Kindle, Seller services, etc.) also use our libraries and APIs to deliver solutions to their own customers. We are a full stack shop, and our team capabilities cover the whole solution spectrum, ranging across applied science, large scale engineering services, product management, UX design, and mobile app development for iOS and Android.
US, MN, Minneapolis
AWS Central Economics is an interdisciplinary team on the cutting edge of economics, statistical analysis, and machine learning whose mission is to solve problems that have high risk with abnormally high returns. Our team leverages the strengths of our scientists to build solutions for some of the toughest business problems here at Amazon AWS. We are looking for an exceptionally talented, seasoned, and motivated Economist to manage a team of economists and data scientists to drive the science for AWS. Key job responsibilities Manage a team of economists and data scientists to deliver actionable economic analyses to business leaders, provide leadership on the economics and science used in the analyses, and engage with business leaders to identify challenges AWS faces that call for in-depth economic analyses and to ensure the analyses have their intended impact.
LU, Luxembourg
&ltHire Relocation Requisition - not for posting> Provides insights to leadership on improving Supply Chain cost and Speed by using Data Science and Analytics techniques. Build Dashboards and models to industrialize these findings at scale.
US, VA, Arlington
The People eXperience and Technology Central Science Team (PXTCS) uses economics, behavioral science, statistics, and machine learning to proactively identify mechanisms and process improvements which simultaneously improve Amazon and the lives, wellbeing, and the value of work to Amazonians. We are an interdisciplinary team that combines the talents of science and engineering to develop and deliver solutions that measurably achieve this goal. We are looking for economists who are able to work with business partners to hone complex problems into specific, scientific questions, and test those questions to generate insights. The ideal candidate will work with engineers and computer scientists to estimate models and algorithms on large scale data, design pilots and measure their impact, and transform successful prototypes into improved policies and programs at scale. We are looking for creative thinkers who can combine a strong technical economic toolbox with a desire to learn from other disciplines, and who know how to execute and deliver on big ideas as part of an interdisciplinary technical team. Ideal candidates will work closely with business partners to develop science that solves the most important business challenges. They will work in a team setting with individuals from diverse disciplines and backgrounds. They will serve as an ambassador for science and a scientific resource for business teams, so that scientific processes permeate throughout the HR organization to the benefit of Amazonians and Amazon. Ideal candidates will own the data analysis, modeling, and experimentation that is necessary for estimating and validating models. They will work closely with engineering teams to develop scalable data resources to support rapid insights, and take successful models and findings into production as new products and services. They will be customer-centric and will communicate scientific approaches and findings to business leaders, listening to and incorporate their feedback, and delivering successful scientific solutions. Key job responsibilities Use causal inference methods to evaluate the impact of policies on employee outcomes. Examine how external labor market and economic conditions impact Amazon's ability to hire and retain talent. Use scientifically rigorous methods to develop and recommend career paths for employees. A day in the life Work with teammates to apply economic methods to business problems. This might include identifying the appropriate research questions, writing code to implement a DID analysis or estimate a structural model, or writing and presenting a document with findings to business leaders. Our economists also collaborate with partner teams throughout the process, from understanding their challenges, to developing a research agenda that will address those challenges, to help them implement solutions. About the team We are a multidisciplinary team that combines the talents of science and engineering to develop innovative solutions to make Amazon Earth's Best Employer.
US, WA, Seattle
The People eXperience and Technology Central Science Team (PXTCS) uses economics, behavioral science, statistics, and machine learning to proactively identify mechanisms and process improvements which simultaneously improve Amazon and the lives, wellbeing, and the value of work to Amazonians. We are an interdisciplinary team that combines the talents of science and engineering to develop and deliver solutions that measurably achieve this goal. We are looking for economists who are able to apply economic methods to address business problems. The ideal candidate will work with engineers and computer scientists to estimate models and algorithms on large scale data, design pilots and measure their impact, and transform successful prototypes into improved policies and programs at scale. We are looking for creative thinkers who can combine a strong technical economic toolbox with a desire to learn from other disciplines, and who know how to execute and deliver on big ideas as part of an interdisciplinary technical team. Ideal candidates will work in a team setting with individuals from diverse disciplines and backgrounds. They will work with teammates to develop scientific models and conduct the data analysis, modeling, and experimentation that is necessary for estimating and validating models. They will work closely with engineering teams to develop scalable data resources to support rapid insights, and take successful models and findings into production as new products and services. They will be customer-centric and will communicate scientific approaches and findings to business leaders, listening to and incorporate their feedback, and delivering successful scientific solutions. Key job responsibilities Use causal inference methods to evaluate the impact of policies on employee outcomes. Examine how external labor market and economic conditions impact Amazon's ability to hire and retain talent. Use scientifically rigorous methods to develop and recommend career paths for employees. A day in the life Work with teammates to apply economic methods to business problems. This might include identifying the appropriate research questions, writing code to implement a DID analysis or estimate a structural model, or writing and presenting a document with findings to business leaders. Our economists also collaborate with partner teams throughout the process, from understanding their challenges, to developing a research agenda that will address those challenges, to help them implement solutions. About the team We are a multidisciplinary team that combines the talents of science and engineering to develop innovative solutions to make Amazon Earth's Best Employer.
US, WA, Seattle
Amazon is looking for talented Postdoctoral Scientists to join our global Science teams for a one-year, full-time research position. Postdoctoral Scientists will innovate as members of Amazon’s key global Science teams, including: AWS, Alexa AI, Alexa Shopping, Amazon Style, CoreAI, Last Mile, and Supply Chain Optimization Technologies. Postdoctoral Scientists will join one of may central, global science teams focused on solving research-intense business problems by leveraging Machine Learning, Econometrics, Statistics, and Data Science. Postdoctoral Scientists will work at the intersection of ML and systems to solve practical data driven optimization problems at Amazon scale. Postdocs will raise the scientific bar across Amazon by diving deep into exploratory areas of research to enhance the customer experience and improve efficiencies. Please note: This posting is one of several Amazon Postdoctoral Scientist postings. Please only apply to a maximum of 2 Amazon Postdoctoral Scientist postings that are relevant to your technical field and subject matter expertise. Key job responsibilities * Work closely with a senior science advisor, collaborate with other scientists and engineers, and be part of Amazon’s vibrant and diverse global science community. * Publish your innovation in top-tier academic venues and hone your presentation skills. * Be inspired by challenges and opportunities to invent cutting-edge techniques in your area(s) of expertise.