Intel and AWS
Amazon Web Services (AWS) and Intel have a more than 16-year relationship dedicated to developing, building, and supporting cloud services. This collaboration helps manage cost and complexity, accelerates business outcomes, and enables scaling to meet current and future computing requirements.
Leveraging the Power of HPC for Social Good
Intel & AWS: Underpinning Solutions to the World’s Most Urgent Issues
AWS and Intel are committed to underpinning solutions to the world’s most critical social and environmental challenges through the power of HPC in the cloud. Intel and AWS have collaborated to bring best-in-class resources and results to partners and customers around the world that are driving impact for good.
Why Intel?
Better Performance per Cost
Intel® technologies, including Intel® Xeon® Scalable processors, deliver exceptional performance per cost in AWS cloud instances.
Hardware-Enabled Security Features
Embedded security capabilities all the way down to the silicon help protect customer workloads on multitenant AWS cloud instances. Intel® technologies with hardware-enabled encryption help isolate and protect data at rest, in transit, and in processing.
The Latest Technology
Intel partners closely with AWS to deploy the latest technologies and software with high levels of optimization. Customers benefit from world-class cloud architecture, workload-tuned instances, and built-in AI and security accelerators as part of the services they buy.
Cloud Capabilities
Intel and AWS provide several jointly developed guides and resources that help customers navigate cloud capabilities, from enterprise applications to IoT, HPC, and AI.
Enterprise Applications
Benefit from our fast pace of innovation, broad and deep functionality, and proven operational experience powered by Intel.
Learn more
Compute at the Edge
AWS offers organizations a fast, secure path to cloud while enabling customers to run applications on the same Intel® technology at the edge and in the cloud.
Download the presentation
High Performance Computing
Together AWS and Intel offer customers a broad and deep set of leading compute, networking.
Read the article
Artificial Intelligence and Machine Learning
Amazon EC2 instances, with Intel performance optimizations, enable customers to create intelligent and innovative new products and experiences.
Download the Enterprise AI guide
Performance for Every Workload
Database Workloads
OpenFOAM Performance on AWS
3rd Gen Intel® Xeon® Scalable processors provide significant performance gains for OpenFOAM running on AWS.
STREAM on AWS M6i vs. M6a
Get up to 2.75 times the STREAM Triad performance with Amazon EC2 m6i instances vs. m6a instances with AMD EPYC processors.
PostgreSQL on AWS M6i vs. M5
Compared to m5 instances with older Intel® processors, m6i instances delivered consistently stronger PostgreSQL performance.
PostgreSQL on AWS M6i vs. M6a
See up to 24% more NOPM with m6i instances with 3rd Gen Intel® Xeon® Scalable CPUs vs. m6a instances with 3rd Gen AMD EPYC CPUs.
Improve MariaDB Performance 1.29x on AWS EC2 M6i
New EC2 M6i instances featuring 3rd Gen Intel® Xeon® Scalable processors handled more transactions per minute.
Get Microsoft SQL Data 1.37x as Fast with AWS
Speed up data analysis with AWS M6i instances featuring 3rd Gen Intel® Xeon® Scalable processors.
Book More Reservations at Once on MySQL with AWS
Get better MySQL transactional database performance with AWS M6i instances featuring 3rd Gen Intel® Xeon® Scalable processors.
Support Up to 1.38x MySQL eCommerce on AWS M6i
Handle more transactions per minute with AWS M6i instances featuring 3rd Gen Intel® Xeon® Scalable processors.
VMware Cloud Hosts 2x VMs on AWS i3en
Improve scale-out performance with i3en instance types of VMware Cloud on AWS on 2nd Gen Intel® Xeon® Scalable processors.
AWS EC2 M5 VM Did Up to 1.67x the Work of EC2 M6G
Reap greater performance by selecting M5 instances featuring 2nd Gen Intel® Xeon® Scalable processors.
Increase MySQL by 1.65x with AWS EC2 M5 Instances
Reap 1.32x the performance per dollar with AWS M5 instances featuring 2nd Gen Intel® Xeon® Scalable processors.
Downsize AWS from 64 to 32 vCPUs and Save Money
Cut costs by choosing R5n instances featuring 2nd Gen Intel® Xeon® Scalable processors for your MySQL database work.
Intel® Xeon® Processor VM Achieved 1.2x as Java Web
Get greater performance by selecting instances featuring 2nd Generation Intel® Xeon® Scalable processors.
Handle 1.45x More PostgreSQL Transactions on AWS
Boost database performance with newer C5 instances featuring 2nd Gen Intel® Xeon® Scalable processors.
Process 1.38x More PostgreSQL Transactions on AWS
Boost database performance with newer R5 instances featuring 2nd Gen Intel® Xeon® Scalable processors.
AWS Expands SAP Portfolio with New VM Instances
Take advantage of improved features and SAP NetWeaver application certification with new M6i instances.
Process Up to 1.18x More MySQL with AWS EC2 C5
Improve performance per dDollar by 1.06x with AWS C5 instances featuring 2nd gen Intel® Xeon® scalable processors.
Analytics Workloads
GROMACS Performance on Intel Instances on AWS
3rd Gen Intel® Xeon® Scalable processors provide significant performance gains for the GROMACS workload.
Databricks with Photon on AWS Instances
Across AWS instances with various processor generations, enabling photon improved decision support workload performance.
Databricks on AWS R5d Instances vs. R5a Instances
With photon enabled, R5d instances delivered better decision support performance and value than R5a instances with AMD EPYC.
Databricks on AWS R5d Instances vs. R6gd Instances
With photon enabled, R5d instances delivered better decision support performance & value than R6gd instances with graviton2.
Image Classification on AWS M6i Instances vs. M5n
AWS M6i instances with 3rd Gen Intel® Xeon® scalable processors accelerated image classification over M5n instances.
Wide & Deep Recommender on Amazon M6i vs. M6a
AWS M6i instances performed more inference operations per second than m6a instances with 3rd Gen AMD EPYC processors.
Wide & Deep Recommender on Amazon M6i vs. M5n
Across different instance sizes, AWS M6i instances performed more inference operations per second than M5n instances.
Speed Splunk by 1.93x with Intel® Xeon® Processors
Get reports from data and act on insights faster with C5 instances featuring 2nd Gen Intel® Xeon® Scalable processors.
Better Performance and Container Scaling on Splunk
Bare-metal clusters with 3rd Gen Intel® Xeon® Scalable processors improve Splunk indexing, searching, and container scaling.
Boost Splunk Up to 5.13x with AWS EC2 C5 Instances
Gain insight from machine data faster with C5 instances featuring 2nd Gen Intel® Xeon® Scalable processors.
AWS EC2 R5n Supported Nearly 2x Kubernetes Users
Get more from your AWS investment by selecting newer R5n instances featuring 2nd Gen Intel® Xeon® Scalable Processors.
AWS EC2 M5 Achieved 1.2x Java Web Services Work
Get greater performance by selecting M5 instances featuring 2nd Gen Intel® Xeon® Scalable processors.
Handle Up to 1.70x More ASP.NET Work on AWS
Get up to 1.36x more ASP.NET performance per dollar with AWS M5 instances featuring 2nd Gen Intel® Xeon® Scalable processors.
Do Up to 1.7x More ASP.NET Work on AWS
Improve ASP.NET performance with AWS M5 instances featuring 2nd Gen Intel® Xeon® Scalable processors.
BERT-Large Inference on AWS M6i vs. M6g
M6i instances performed more inference work than M6g instances featuring AWS Graviton2 processors.
BERT-Large Inference on AWS M6i vs. M5n
M6i instances performed more inference operations per second than M5n instances with 2nd Gen Intel® Xeon® Scalable processors.
Image Classification on AWS M6i Instances vs. M5n
AWS M6i instances with 3rd Gen Intel® Xeon® Scalable processors accelerated image classification over M5n instances.
Wide & Deep Recommender on Amazon M6i vs. M6a
AWS M6i instances performed more inference operations per second than M6a instances with 3rd Gen AMD EPYC processors.
Wide & Deep Recommender on Amazon M6i vs. M5n
Across different instance sizes, AWS M6i instances performed more inference operations per second than M5n instances.
Classify 6.17x More Images per Second with AWS
Boost ResNet50 inference performance with AWS M5n instances featuring 2nd Gen Intel® Xeon® Scalable processors.
Get Recommendations from 2.94x Faster with AWS
Improve wide & deep inference performance with AWS M5n instances featuring 2nd Gen Intel® Xeon® Scalable processors.
Image Classification on AWS M6i Instances vs. M6a
AWS M6i instances with 3rd Gen Intel® Xeon® Scalable processors accelerated image classification over M6a instances.
Improve TensorFlow* Performance on AWS Instances by Leveraging Intel® Deep Learning Optimizations
Intel® Optimizations for TensorFlow* with the oneDNN library has been available in the official TensorFlow release since version 2.5. By default, the oneDNN optimizations are turned off before version 2.9. You can enable the oneDNN optimizations by setting the environment variable TF_ENABLE_ONEDNN_OPS to 1. When enabled and configured properly, these oneDNN optimizations can provide a 3X performance improvement over using the stock TensorFlow*.
Out-of-Box Performance Improvement from TensorFlow* v2.9 on AWS 3rd Gen Intel® Xeon Instance Type and Performance Comparison with AWS 3rd Gen AMD EPYC Instance Type
The oneDNN optimizations from TensorFlow v2.9 can provide up to a 3X out-of-box performance improvement over v2.8 where you would need to enable these optimizations manually to see similar performance improvements. Moreover, among different CPU instance types such as C6a with AMD EPYC and different TensorFlow optimization such as ZenDNN, the best performances from our benchmarking are achieved with Intel C6i instance type from TensorFlow* v2.9.
STREAM on AWS M6i vs. M5
Amazon EC2 m6i instances offered up to 53% better STREAM triad performance vs. m5 instances.
SPECrate 2017 Floating Point on AWS M6i vs. M6a
Achieve 58% better estimated SPECrate®2017_fp_base performance than m6a instances with 3rd Gen AMD EPYC™ processors.
SPECrate 2017 Floating Point on AWS M6i vs. M5
AWS m6i instances with 3rd Gen Intel® Xeon® Scalable CPUs beat m5 instances with older Intel CPUs in Floating-Point performance.
SPECrate2017_int_base on AWS m6i vs. m5
m6i instances offered better estimated SPECrate®2017_int_base scores than m5 instances.
Kafka on AWS R5 vs. R4 and C5 Instances
R5 instances outperformed both R4 instances and C5 instances with the same or previous-generation processors Kafka on AWS R5 vs. R4 and C5 instances.
Instance Types
AWS enabled by Intel offers a broad selection of instance types to help customers match their workloads to the best-fit combination of architecture and technologies.
General Purpose
Intel-based general purpose AWS instances offer balanced cloud infrastructure to suit a wide variety of workloads. These instances are ideal for applications that use compute, memory, and networking in equal proportions, such as web servers and code repositories.
Learn more about general purpose instances ›
- M6i: Intel® Xeon® 3rd gen Scalable processors | Ice Lake
- M4: Intel® Xeon® E5-2686 processors or Intel® Xeon® E5-2676 processors | Broadwell and Haswell
- M5: Intel® Xeon® Platinum 8175M processors | Skylake
- M5n, M5zn: Intel® Xeon® Scalable processors | Cascade Lake
- T2 (burstable): Intel® Xeon® processors | Broadwell and Haswell
- T3 (burstable): Intel® Xeon® Scalable processors | Skylake
Compute Optimized
Compute optimized AWS instances enabled by Intel focus on delivering high performance for compute-intensive workloads, like media transcoding, HPC, scientific modeling, and high-traffic servers, including gaming services.
Learn more about compute optimized instances ›
- C6i: Intel® Xeon® 3rd gen Scalable processors | Ice Lake
- C4: Intel® Xeon® E5-2666 processors | Haswell
- C5: Intel® Xeon® Scalable processors | Cascade Lake
- C5n: Intel® Xeon® Platinum 8124M processors | Skylake
Accelerated Compute
These Intel-enabled AWS instances combine powerful CPUs with other add-in accelerators and GPUs to support workloads with intensive floating-point calculations, graphics processing, or data pattern matching.
Learn more about accelerated compute instances ›
- F1: Intel® Xeon® E5-2686 v4 processors | Broadwell
- G3: Intel® Xeon® E5-2686 processors | Broadwell
- G4: Intel® Xeon® Scalable processors | Cascade Lake
- P2: Intel® Xeon® E5-2686 processors | Broadwell
- P3: Intel® Xeon® E5-2686 v4 or P-8175M processors | Broadwell or Skylake
Memory Optimized
For workloads that process large data sets in memory, AWS memory optimized instances enabled by Intel prioritize memory capacity and input/output speed.
Learn more about memory optimized instances ›
- R6i: Intel® Xeon® 3rd gen Scalable processors | Ice Lake
- High memory: Intel® Xeon® Platinum 8176M or Scalable processors | Skylake or Cascade Lake
- R4: Intel® Xeon® E5-2686 processors | Broadwell
- R5: Intel® Xeon® Platinum 8175 processors | Skylake
- R5n, R5b: Intel® Xeon® Scalable processors | Cascade Lake
- X1e/X1: Intel® Xeon® E7 8880 v3 processors | Haswell
- Z1d: Intel® Xeon® Platinum 8151 processors | Skylake
Storage Optimized
Intel-enabled storage optimized instances focus on delivering high-capacity data storage with high sequential read/write speeds. This allows for low latency data transmission to other applications.
Learn more about storage optimized instances ›
- D2: Intel® Xeon® E5-2676 processors | Haswell
- D3, D3en: Intel® Xeon® Scalable processors | Cascade Lake
- H1: Intel® Xeon® E5-2686 processors | Broadwell
- i3: Intel® Xeon® E5-2686 v4 processors | Broadwell
- i3en: Intel® Xeon® Scalable processors | Skylake
Customer Success Stories
These case studies illustrate how Intel and AWS have helped organizations achieve real-world success with fast and efficient services delivery.
Career Launcher: Educators Launch Online Learning
In the face of the pandemic lockdown, Career Launcher successfully delivers learning on a massive scale in India.
MicroSeismic Evaluates HPC in the Cloud
Intel® processor-based instances in the cloud showed potential for cost savings, faster processing, and business opportunity.
DataCubes: Reducing Insurance Underwriting Risks
Using AI, commercial property and casualty underwriters aim to be the first to quote while reducing risk.
Astera Labs: Cloud-Based HPC for Better Designs
Quick access to the latest custom Intel® Xeon® Scalable processors in Amazon EC2 Z1d instances proves invaluable.
Resources
Continue your AWS cloud journey by learning about unique services and events.
Attend a Virtual Event
Explore Products and Services
AWS Wavelength: Ultra-low-latency apps for 5G