The Intel team is gearing up for a fantastic re:Invent later this month, and as I reviewed the plan today, I could see the results of the decade+ strategic collaborations between Intel and Amazon Web Services (AWS). It is through this deep technology relationship that we continue to build trust between the two companies, expand the breadth and depth of our joint innovations, and bring reliable, consistent, performant computing solutions to every single AWS customer.
An important milestone in the history of cloud computing was when AWS launched its first Elastic Compute Cloud (EC2) instance in 2006. Starting in 2008 the m1.small, ran on Intel® Xeon® processors. From 2008 to 2018, the joint innovation accelerated as AWS customers bring along more varieties of workloads to the cloud. With the current generation of Intel® Xeon® Scalable processors, AWS has been able to create the largest number of new and differentiated instances based on heavy engineering collaboration around CPU customization and software optimization. I love numbers – they speak for themselves. Let’s look at a few:
- The vast majority of instances have been launched by AWS and all are running Intel Xeon processors spanning 7 generations, including Nehalem, Westmere, Sandy Bridge, Ivy Bridge, Haswell, Broadwell, and our most current generation - Skylake
- 54 different instances are based on the latest Intel Xeon Scalable processor (Skylake), out of which 51 are built upon custom processors tailored to improve perf/TCO for AWS users’ workloads
- Intel Xeon cpus are available across all 18 AWS regions, 55 availability zones serving hundreds of customers and workloads reliably year in and year out
What else do these mean to AWS customers? Businesses large and small are seeing several benefits:
Scalable, Customizable Performance
The breadth of the Intel-based EC2 instance family and their ready availability give customers flexibility to choose the most efficient solution across a diverse set of workloads and usages such as:
- Compute Optimized (C instances)
- General Purpose (M and T instances)
- Memory Optimized (R, X and Z1d instances)
- Accelerated Computing (P, G, and F instances)
- Storage Optimized (H,I and D instances)
With the wide availability of Intel Xeon processors through AWS, customers enjoy a consistent computing architecture across the full spectrum of performance, power, and price for reliable application delivery from data center to edge, without needing to worry about compatibility issues when making infrastructure decisions. Clouds built on Intel architecture deliver maximum workload portability while easing deployment complexity by providing a uniform application environment.
Deployment Simplicity and Agility
AWS and Intel have jointly serviced hundreds of real-world workloads together and have worked hard to optimize infrastructure hardware and software for these workloads based on their diverse characteristics. These optimizations make our customers’ life much easier and their cloud deployment much faster. Let’s take the new z1d instance as an example. z1d represents rich over a year of co-engineering engagement with AWS to develop a customized Intel Xeon Scalable processor running up to 4.0 GHz, powered by sustained all-core Turbo Boost which is perfect for Electronic Design Automation, financial simulation, relational database, and gaming workloads. Customers running these workloads can quickly deploy their applications with the confidence that they get high performance efficiency and bang for their buck.
In addition, cloud customers benefit from powerful application optimizations enabled for Intel architecture by the industry’s broadest ISV ecosystem, such as Cloudera, Microsoft, SAP, VMware, and open source communities such as Apache Foundation, Cloud Native Computing Foundation, Linux Foundation, MXNet, and Tensorflow to name a few. Developers can build their applications once and run them on any Intel based on-or-off premise cloud with minimal rework for ultimate agility.
Ease of Mind with Trusted Reliability
Clouds built on Intel architecture deliver the most predictable, industry-leading per core performance and workload acceleration. With Intel’s rich experience working with enterprises on mission-critical business infrastructure and meeting the stringent service level agreements (SLA) of business applications, Intel Xeon processors boast industry-leading server uptime and reliability.
So much more to say but we’ll have to wait until re:Invent where we have some big announcements ready to go. For more information around our collaboration with AWS and the value we bring, visit the Intel page on AWS site: http://aws.amazon.com/intel/.
Intel and AWS have enjoyed a strong engineering relationship since their founding, working together to better serve AWS’s customers. As the technology landscape evolves, our commitment to build deep technology and sales collaborations with our leading cloud customers remains steadfast. I’m really looking forward to showing off the latest Intel technology at re:Invent, and how you can take advantage of it through AWS.
See you in Las Vegas at re:Invent 2018!