Hey Reader
Welcome to Issue # 10 of AWS Graviton Weekly, which will be focused on sharing everything that happened in the past week related to AWS Graviton: from November 4th, 2022 to November, 11th, 2022.
Before continuing, I want to thank the AWS team, especially Will Koplitz for sharing this incredible lab, where you can identify which RDS instances can be upgraded to Graviton2-based instances.
This is perfect for companies and teams using RDS actively, where you could save up to 35% performance improvement and up to 52% price/performance improvement.
And another incredible resource is this workshop created by the one and only Syl Taylor, which focused on how to run confidential Machine Learning workloads on AWS Graviton and AWS Nitro Enclaves.
If you are doing this kind of work, definitely you must check it out.
Time to save big, folks.
This issue is sponsored by Kit
Kit is the incredible platform I use to send this email to you every single week.
You can segment your audience, you can build incredible email sequences (game changer !!!) and more cool stuff.
It's very simple and intuitive to use and it's totally free to start with it.
If you are planing to launch your own newsletter, Kit is the email marketing platform to be.
News
Articles and Tutorials
Run ML inference on EC2 (Graviton) using Apache TVM, by Syl Taylor (Specialist Solutions Architect - EC2 Graviton @ AWS )
Simplifying Amazon EC2 instance type flexibility with new attribute-based instance type selection features, by Rajesh Kesaraju (Sr. Solution Architect, EC2-Flexible Compute at AWS) and Peter Manastyrny (Sr. Product Manager, EC2 at AWS)
Carlos Manzanedo Rueda (a WW Solutions Architecture Leader, Flexible Compute - EC2 Spot at Amazon Web Services) described this perfectly here:
Yet another release that will help customer simplify their configurations. If you have not heard about Attribute Based Instance Selection (ABS) you'd like this one.
ABS lets you express your instance requirements in Auto Scaling Group, Fleet, Terraform, etc as a set of attributes, such as vCPU, memory, and storage. Your requirements are translated by ABS to all matching instance types, simplifying the creation and maintenance of instance type configurations.
With this release Attribute based instance selection is becoming more powerful and easier to use. New functionality has been released to include Network Bandwidth and an AllowedInstanceTypes that complements ExcludedInstanceTypes attribute.
Coupang Rocket Growth's ML platform: service more than 20 models and cost-effective traffic processing, by Harsha Konda (Staff Engineer at Coupang). The original link is in Korean, so I used Google Translate to have an English version. Download it here
changelogs.md, a project created by Nathan Peck (a Developer Advocate from AWS focused on Containers and Serverless stuff) recently updated this project, moving to Graviton. You can check the code here
Introducing the AWS Lambda Telemetry API, by Anton Aleksandrov (Principal Solution Architect at AWS) and Shridhar Pandey ( Senior Product Manager at AWS)
The Road to Graviton, by Rivergon
Cloud Optimization Best Practices for AWS, Azure, and GCP, by Keren Shmuely (Marketing Director at Granulate)
Slides, Videos and Audio
[VIDEO] Graviton Workloads | The Keys to AWS Optimization | S5 E8 with Will Koplitz (Senior Specialist Solutions Architect - Graviton - ISVs at AWS), John Masci (AWS Strategic Accounts at AWS) and Rem Baumann (Commercial Architect at AWS). Highly recommended
[VIDEO] Next Generation Infrastructure - Optimizing Efficiency & Costs (Twitter Space Recap), with Randall Hunt (VP of Cloud Strategy and Solutions at Caylent) and Corey Quinn (Chief Cloud Economist at The Duckbill Group), where they discussed how Graviton offers some of the most power efficient compute available.
[VIDEO] Automate Your API Life Cycle Management with Kong and AWS, by Anuj Sharma (Container Specialist Solution Architect at AWS) and Claudio Acquaviva (Software Architect at Kong), where they explained how using Graviton instances to optimize costs up to 40%
[VIDEO] AWS Container Day: Cut your cluster costs - How to monitor and reduce your compute costs, by Vipin Mohan (Principal Product Manager at AWS) and Ajay Tripathy (co-founder and CTO at Kubecost), where they explained how to reduce compute costs in your environment using Karpenter and Graviton
[VIDEO] Operating OpenZFS at Scale, by Satabdi Das (Senior Software Development Engineer at Amazon), where she explained how Amazom FSx is supported by AWS Graviton
Jobs
Events
[MEETUP] AWS Meetup Tashkent #1: Friday, November 11, 2022. 6:30 PM to 9:00 PM UZT at Westminster International University. Mikhail (Mike) Golubev (AWS) and Egor Miasnikov will be talking about Graviton.
Building real-time applications to utilize AWS Graviton: this interesting article from Matt Bushell (Sr. Director of Product Marketing at Aerospike) explained how Fidelity and Wix.com use Aerospike on top of AWS Graviton to build real-time apps. They will give a session at the upcoming reInvent 2022 called Building real-time applications to utilize AWS Graviton (Session ID # PRT246). Speakers: Tal Gabay ( Engineering Manager, Wix.com), Raj Pai (Vice President, AWS), Subbu Iyer (President and CEO, Aerospike), and Kyle Bush( VP, Data Analytics Architecture, Fidelity Investments)
Sustainability in the cloud with Rust and AWS Graviton (Tuesday, November, 29th ), with Emil Lerch (Principal DevOps Specialist at AWS) and Esteban Kuber (Principal Software Engineer at AWS)
AWS Graviton adoption by Honeycomb.io using Amazon EC2 and AWS Lambda, by Sunil Ramachandra (Sr.Solutions Architect at AWS) and Pierre Tessier (Director of Solution Architects at Honeycomb.io) (Monday, November 28th , 4:45 PM - 5:45 PM PST & Tuesday, November 29th , 5:00 PM - 6:00 PM PST )
Quote of the week
We have transitioned 500 ElasticSearch nodes that process 2PB of ingested data per month to R6g instances powered by AWS Graviton2.
With this transition we achieved a 22% price reduction, 20% performance improvement, and 40% lower load on our indexing nodes over comparable fifth generation x86-based instances.
The transition process took 240 working hours and included changes to our CI/CD pipeline and multi-arch Kubernetes image creation.
Overall the transition to AWS Graviton-based instances helps us improve Coralogix’s business and technical KPIs.
We managed to reduce Coralogix COGS (cost of goods sold) and improve the customer experience for running queries on our platform.
As a next step, we plan to expand the use of AWS Graviton-based instances in other parts of our application and adopt the latest AWS Graviton3-based instances to gain even better business and technical benefits.
Oded David, Director of Engineering and Enablement at Coralogix