← Back to list

AWS takes a big bet on Machine Learning and IoT with bare metal instances.

A few years ago, AWS decided that it wanted to modernize the architecture of the EC2 platform. That meant moving the networking and storage…

Abhinay Dronavalli · 2017-11-29 06:46 · 92 claps · 2.0 min read
#aws #awsreinvent #ec2 #containers #machine-learning
Open on Medium ↗
Wiki topics: RAG · RAG & Retrieval ML · Machine Learning EDU · Education & Learning ☁️ · DevOps & Cloud 📟 · Gadgets & IoT 🏛️ · Architecture

AWS takes a big bet on Machine Learning and IoT with bare metal instances.

AWS Announcing bare metal instances at #AWSReinvent2017.

AWS Announcing bare metal instances at #AWSReinvent2017.

A few years ago, AWS decided that it wanted to modernize the architecture of the EC2 platform. That meant moving the networking and storage stack to a new platform which AWS dubbed the “Nitro Architecture”. To do this, AWS started innovating in the area of ASIC (Application Specific Integrated Circuit) along with Annapurna Labs, to provide the performance that matches the application workload as ASICs are customized for a particular use, rather than intended for general-purpose use, and introduces a virtual machine called KVM, hypervisor that ships standard with a Linux core. This is a new architecture of virtualization on EC2 that offloads networking, storage and management tasks from EC2 host servers to dedicated hardware.

Today AWS Announces bare metal instances with i3.metal instance, EC2 instances that offer the best of both worlds, allowing the operating system to run directly on the underlying hardware while still providing access to all of the benefits of the cloud. The instance gives you direct access to the processor and other hardware. AWS is launching these bare metal instances as part of its i3 instance family but we can expect it to bring them to a wider range of instance families over time. These new instances are now going into public preview where developers need to sign up for preview.

With bare metal, AWS users get direct access to the hardware and access to virtually 100 percent of the hardware’s resources without any major overhead. They also allow AWS users to run their own virtualization stacks, which also gives them more control over their cloud servers. Along with this, though, this now also opens up EC2 to applications that users couldn’t previously license for virtualized environment or couldn’t get support for.

Hardware specifications of instance are:

  • Processing — Two Intel Xeon E5–2686 v4 processors running at 2.3 GHz, with a total of 36 hyperthreaded cores (72 logical processors).
  • Memory — 512 GiB.
  • Storage — 15.2 terabytes of local, SSD-based NVMe storage.
  • Network — 25 Gbps of ENA-based enhanced networking.

The bare metal instances will still be able to make use of all the usual EC2 services like Elastic Load Balancing, Auto Scaling, Amazon CloudWatch, Auto Recovery, and so forth. They can also access the full suite of AWS database, IoT, mobile, analytics, artificial intelligence, and security services, along with providing the basic ground level hardware access to machines.

In a nutshell, AWS bare metal instances are,

  1. Great for custom workloads
  2. Specific Hypervisor, KVM
  3. Customer hostile licensing
  4. Near 100% utilization of resources
  5. Direct hardware access with elasticity and scalability
  6. Guarantees consistent performance

The bare metal instances are amazing for running data intense work loads with nosql databases, running containers, deep learning and machine learning.


메타데이터
post_id
a8a07aa32183
slug
aws-takes-a-big-bet-on-machine-learning-and-iot-with-bare-metal-instances-a8a07aa32183
url
https://medium.com/@abhidrona/aws-takes-a-big-bet-on-machine-learning-and-iot-with-bare-metal-instances-a8a07aa32183
canonical_url
https://medium.com/@abhidrona/aws-takes-a-big-bet-on-machine-learning-and-iot-with-bare-metal-instances-a8a07aa32183
author_url
https://medium.com/@abhidrona
status
ok
fetched_at
2026-07-30 08:26:29