StepSecurity, maker of the Harden-Runner tool used by more than 20,000 open-source projects, announced a US$20 million Series A led by Elephant to protect developer machines, package downloads, CI/CD pipelines and agent sandboxes.
Amazon EKS Hybrid Nodes now bills large servers as 32 vCPUs
EditorialBy TrustList Editorial
AWS now bills each Amazon EKS Hybrid Nodes server for at most 32 vCPUs, so at the first pricing tier a 64 vCPU host costs 50% less and a 256 vCPU host 88% less; existing nodes must be removed and reattached to get the new price.
About Amazon EKS Hybrid Nodes now bills large servers as 32 vCPUs
Amazon EKS Hybrid Nodes now bills large servers as 32 vCPUs
18 November 2026: Amazon Web Services has capped what it charges for each server attached to Amazon EKS Hybrid Nodes, the option that lets companies run their own on-premises and edge machines as nodes in an Amazon Elastic Kubernetes Service cluster. Each hybrid node is now billed for at most 32 vCPUs, however many it has. At the first pricing tier, AWS says, a 64 vCPU node now costs 50% less than before and a 256 vCPU node 88% less.
Not yet independently verified. The change, the 50% and 88% figures and the advice on existing nodes come from the AWS announcement posted on 18 November 2026 and the Amazon EKS pricing page, both read on 19 November. The per-vCPU hourly rates on the pricing page are drawn by script and could not be read, so no rate is quoted here. The two percentages are AWS's own figures for the first pricing tier. We will update this when it can be confirmed, and remove this note.
How the charge works now
Hybrid Nodes are charged per vCPU per hour, based on what each node reports to Kubernetes, from the moment it joins a cluster until it is removed. On bare metal with hyperthreading switched on, each physical core reports two vCPUs, which made large servers expensive to attach. Under the new rule a node reporting more than 32 vCPUs is billed as 32; a node with 32 or fewer is still billed for what it has. The existing tiers stay in place, applied to total monthly vCPU-hours in each region, across all accounts in an AWS Organization where consolidated billing is used, so the effective rate still falls as usage grows.
The new price applies automatically to hybrid nodes added from now on. Nodes already attached to a cluster keep the old billing until they are deleted and reattached, AWS says. The change is available in every region that offers EKS Hybrid Nodes.
Who gains
AWS aims the change at large bare-metal and GPU servers running heavy work, such as AI training and inference, that must stay on premises for data residency, latency or regulatory reasons. Until now the per-vCPU charge grew with the size of the machine, which pushed some teams to split big hosts into virtual machines or to keep them outside EKS altogether.
What platform teams should do
- List the hybrid nodes that report more than 32 vCPUs; they are the ones whose bill changes.
- Plan the delete-and-reattach step for those nodes in a maintenance window: drain each node, remove it from the cluster and join it again, checking that daemon sets, storage drivers and GPU operators come back as expected.
- Recheck any cost model that compared EKS Hybrid Nodes with running a separate Kubernetes distribution on premises, since the answer for large GPU hosts may now be different.
- Note that hybrid nodes running on Amazon EC2 instances are also charged the hybrid nodes fee, so the cap matters there too.
Teams should read the current per-vCPU rates on the EKS pricing page before they rework budgets, and confirm the saving on the first full monthly bill after reattaching.
Sources
- AWS: Amazon EKS Hybrid Nodes reduces price for large hosts, 18 November 2026
- Amazon EKS pricing, 19 November 2026
Categories & features
TrustList Weekly
The week in software and IT, in one email
The news that matters to buyers, new rankings and our own research. Every Thursday, free, and easy to leave.
You may be interested in
nginx released mainline 1.31.9 and stable 1.30.7 on 18 November 2026 with fixes for ten vulnerabilities, three rated major: HTTP/2 flow-control memory corruption, a TLS 1.3 request-body loop and a mail proxy use-after-free.
One of the four high-severity bugs let workspace users read Kubernetes Secrets from other namespaces; GitLab.com is already patched and Dedicated customers need do nothing.
The GitLab regex flaw lets an unauthenticated user run code through a CI/CD configuration; the Cisco flaw reloads ASA and FTD firewalls. Both entered the exploited list on 29 October with a three-day deadline.
Robusta, founded in 2021, rebuilt its Kubernetes troubleshooting tool around an AI agent; it is the second company Freshworks has agreed to buy in October, after Torii, and no price has been given.
We classified our 198 security advisories from 22 September to 28 October. 64 described attacks already under way, edge devices led, half of KEV listings came within five days of a fix, and the CVSS score was a weak guide to which ones mattered.
The old app is renamed "GitHub (deprecated)" first. Its replacement is the former GitHub Notifications app, which now lets GitHub Copilot turn a notification thread into a pull request, if admins allow it.
CVE-2026-111497 is rated 9.2 by Google and hits every Brotli release from 1.0.3 to 1.2.0. Envoy, the PyPI packages and Debian have shipped fixes; Ubuntu still ranks it medium.
More on TrustList
Everything here links back to the same verified catalogue. Pick your next stop.
- More DevOpsThe ranking for this subject
- CompaniesAgencies, consultancies and IT service providers, ranked by verified reviews.
- ProductsSoftware and SaaS with pricing, features, integrations and alternatives.
- AwardsAnnual recognition decided by verified reviews and an independent jury.
- LaunchesNew products and releases, voted up by the community every day.
- AI ModelsBenchmark scores and community ratings for every major model.
- RequestsBuyers describe what they need; vendors respond directly.
- PeopleReviewers, authors and makers with public profiles.
- ComparePut up to four listings side by side before you shortlist.