Why using Graviton 2 will make your team better
I recently started using graviton 2 with EKS, in addition to having an operating cost 30% to 40% cheaper than an ec2 and a much better performance, the question always remains why should we move from an x86 architecture to ARM architecture?
The answer is simple cost and performance, but what are the main points that make Graviton 2 so badass and why it will make your team better.
What are the benefits of using AWS Graviton?
The most significant benefits of AWS Graviton are reduced costs, low latency, better scalability, improved availability, and increased security.
1. Economical
The processor family is based on the Arm architecture. That means they are likely to be System on a Chip (SoC). This further translates into lower energy consumption costs, delivering satisfactory performance for most customers.
2. Ecosystem Support
AWS Graviton and Graviton2 are based on the 64-bit Arm core architecture. Several Linux-based operating systems support the configuration. They include Amazon Linux 2, SUSE and Red Hat. This gives customers more choice.
3. Effective CPU Power
AWS Graviton processors also deliver up to 3.45% higher performance over traditional architecture. They also provide more straightforward processor implementations than X86 processors.
4. Built for General
Purpose AWS Graviton cores are also built to improve efficiency in servers, mid-range storage processes, microservices, and cluster computing.
5. Offers scalable workload
They provide users with an extensive set of burstable workload services such as scale microservices, small and medium database services, virtual desktops and a selection of business-friendly applications critics.
6. Build in computer-intensive model
Processors are also built in a computer-intensive model, such as HD video performance computing, video encoding, gaming, and CPU-based computer learning processes.
7. Offers Enhanced Networking
Expect support for a C6gn network with Elastic Fabric Operator (EFO) 100 Gbps network capabilities.
I’ve been working with high performance teams for a long time and it was still nice and fun to see how a processing architecture migration went so smoothly.
Okay, we use Flutter and Java, which are already languages prepared to run on a multi-cpu architecture, even so I confess that the literature on which packages to use is still not very clear, but I already leave here for you the two dockerFiles ready to use on eks with graviton 2 :)
https://gist.github.com/randomk/66e0ddfe5cbe0ba2031458a38fbeee0a
https://gist.github.com/randomk/c9cc17ffc716b55ac4c6f4e4997ad875
And why will your team get better?
- Creating multi-cpu images makes the team stronger, with the advent of the docker world everyone got a little lazy about the x86 world, but with the popularity of raspberry pi, new macbook m1 and graviton 2 and 3 it is more and more necessary to think about multi-cpu environments
- Using new technologies allows us to review basic concepts that have already been forgotten, in the background everything is linux it is always very important to know which technologies and libs work and why.
- Thinking about the cost, today I work in a very strong multidisciplinary team, but everyone is always very concerned about the cost, we always have a monthly meeting to evaluate everything that has been done and see if we can improve or make something cheaper.
- Always keep evaluating the new cloud technologies where you are running your applications, in our case we use AWS as all our production is already running with graviton 2, I believe that the migration to graviton 3 will be very smooth :)
I hope you liked it and please evaluate the migration to graviton 2 or 3 , we are very happy with the performance and cost
Thank you all for your help!
Nilton Arthur Vitor Felipe Augusto Fonseca Barbosa Icaro
Join FAUN: Website 💻|Podcast 🎙️|Twitter 🐦|Facebook 👥|Instagram 📷|Facebook Group 🗣️|Linkedin Group 💬| Slack 📱|Cloud Native News 📰|More.
If this post was helpful, please click the clap 👏 button below a few times to show your support for the author 👇
