Announcement

November 19, 2026

Query PerfectScale from your terminal or AI agent

Look up clusters, workloads and costs from the PerfectScale CLI. Create a token in the new API Tokens settings.

Shipped by

Bogdan Kirylyuk, Abror Kenjaev, Anton Kalashnikau
Query PerfectScale from your terminal or AI agent
Query PerfectScale from your terminal or AI agent

Announcement

November 13, 2026

HPA Automation: PerfectScale optimizes HPA settings

PerfectScale now finds and fixes HPA settings that drive up costs, like oversized minReplicas or overly low utilization targets. You stay in control by choosing which workloads to automate. Review findings in PodFit, enable automation where you want it, and track every change in the Automation Audit log.

Shipped by

Kostas, Dimitris, Alexandra, Karim, Evgenii
HPA Automation: PerfectScale optimizes HPA settings
HPA Automation: PerfectScale optimizes HPA settings

Announcement

November 13, 2026

Karpenter Automation: Unblock Node Consolidation

PerfectScale now detects the Karpenter NodePool misconfigurations that keep underutilized nodes from consolidating.

Shipped by

Aleksandr, Alexandra, Amit, Mireya, Tania, Evgenii
Karpenter Automation: Unblock Node Consolidation
Karpenter Automation: Unblock Node Consolidation

Announcement

November 13, 2026

Binpacking Automation: Drain Blocked Nodes and Actually Cut Your Bill

Workload automation shrinks the pods, but the cluster bill does not go down. The pods got smaller, the nodes stayed. One pod that refuses to move keeps each half-empty node alive. PerfectScale now proves the pods fit elsewhere, moves them, and removes the node.

Shipped by

Andrii, Amit, Yurii, Abror, Mireya, Tania, Jacek.
Binpacking Automation: Drain Blocked Nodes and Actually Cut Your Bill
Binpacking Automation: Drain Blocked Nodes and Actually Cut Your Bill

Announcement

November 5, 2026

See which workloads PerfectScale automation saved you money on

Realized Savings now traces your savings from the organization total down to every cluster, namespace, and workload

Shipped by

Oleksii, Alexandra, Oleg, Ivan, Itai, Ira, Tania, Eli, Karim, Dmytro
See which workloads PerfectScale automation saved you money on
See which workloads PerfectScale automation saved you money on

Announcement

October 29, 2026

Your HPA settings and pod rightsizing, working together

PerfectScale now analyzes your HPA configurations alongside real workload behavior. You get clear recommendations to tune minimum replica counts, utilization targets, and more. The result is fewer unnecessary replicas and more efficient scaling from the autoscaling setup you already have.

Shipped by

Kostas Messis, Dimitris Kanoutas, Alexandra Makarenko, Mireya Díaz, Ira Chernous, Karim Shakirov
Your HPA settings and pod rightsizing, working together
Your HPA settings and pod rightsizing, working together

Announcement

October 21, 2026

New Bin Packing view: see what's blocking node consolidation

The new Bin Packing view shows how workloads are scheduled across your nodes and what prevents consolidation. Find it in one click from the navigation bar. Clear recommendations help you reduce wasted capacity and cut infrastructure costs.

Shipped by

Eli Birger, Pavel Kostychev, Mireya Díaz, Anton Kalashnikau, Ira Chernous, Abror Kenjaev
New Bin Packing view: see what's blocking node consolidation
New Bin Packing view: see what's blocking node consolidation

Announcement

October 14, 2026

See what PerfectScale automation actually saved you

Realized Savings shows the dollars saved, CPU and memory reclaimed, and resilience issues fixed by PerfectScale automation across your organization (available in Beta).

Shipped by

Oleksii, Alexandra, Oleg, Ivan, Itai, Ira, Eli, Dmytro
See what PerfectScale automation actually saved you
See what PerfectScale automation actually saved you

Announcement

October 1, 2026

Recommendations now respect container-level HPA targets

PerfectScale now recognizes HPA triggers that target a single container, keeping right-sizing recommendations accurate for workloads using container-scoped autoscaling. Recommendations apply only to the container the HPA watches, preventing over- or under-sized suggestions. Podfit and Zoom-In now distinguish pod-level from container-level triggers, so the data you see matches how your autoscaler actually behaves.

Shipped by

Iryna Bohatchenko, Alexandra Makarenko, Mireya Díaz, Karim Shakirov
Recommendations now respect container-level HPA targets
Recommendations now respect container-level HPA targets

Announcement

September 29, 2026

New navigation: find any feature or cluster faster

A clearer menu that shows the full PerfectScale feature set, with organization and cluster pages in separate sections and quick access to any cluster.

Shipped by

Eli, Mireya, Pavel, Abror, Andrei, Ira, Tania
New navigation: find any feature or cluster faster
New navigation: find any feature or cluster faster

Announcement

September 8, 2026

See how much PerfectScale saves you - in EUR, GBP, JPY, and more

PerfectScale now shows cluster costs, pod waste, and potential savings in EUR, GBP, JPY, ILS, CHF, and more.

Shipped by

Idan Gazit, Anton Kalashnikau, Andrii Koval, Michael Sklyar, Alexandra Makarenko, Abror Kenjayev
See how much PerfectScale saves you - in EUR, GBP, JPY, and more
See how much PerfectScale saves you - in EUR, GBP, JPY, and more

Announcement

September 1, 2026

Suspected CPU Runaway Detection is available

PerfectScale now flags containers whose CPU allocations keep climbing without resolving resource pressure, a sign of a workload problem rather than real demand. Automation pauses further CPU increases once your configured thresholds are reached. Set daily and weekly increase limits in your automation YAML to keep rightsizing safe and predictable.

Suspected CPU Runaway Detection is available
Suspected CPU Runaway Detection is available

Announcement

August 20, 2026

Find and fix pods that block node scale-down — via API

Unevictable pods block node scale-down and drain, quietly limiting cluster elasticity and savings. Now visible and fixable via API.

Shipped by

Amit, Tania, Ira, Jacek
Find and fix pods that block node scale-down — via API
Find and fix pods that block node scale-down — via API

Announcement

August 20, 2026

Full-mode ResourceQuota support: increase workload resources up to quota caps

Make full use of resources in namespaces governed by ResourceQuotas

Shipped by

Max Iemelianov, Yevhen Sylenko
Full-mode ResourceQuota support: increase workload resources up to quota caps
Full-mode ResourceQuota support: increase workload resources up to quota caps

Announcement

August 20, 2026

InfraFit Public API

Public API for node utilization, efficiency data, and infrastructure optimization insights

Shipped by

Amit, Evgenii, Dmytro, Idan, Iryna, Tania, Gia, Anton
InfraFit Public API
InfraFit Public API

Announcement

August 20, 2026

Filter workloads by JVM runtime

Find every Java workload in a cluster in one click, no labels required

Shipped by

Oleksii Koshkin, Abror Kenjaev, Vadim Solovey
Filter workloads by JVM runtime
Filter workloads by JVM runtime

Improvement

August 19, 2026

A perfect scale you can actually read

Numbers across PerfectScale now display the way your country writes them.

Shipped by

Vadim Solovey, Andrei Shkoda
A perfect scale you can actually read
A perfect scale you can actually read

Announcement

July 27, 2026

Unevictable pods recommendations in PerfectScale

Find and fix the configurations blocking node scale-down.

Shipped by

Amit, Abror, Andrei, Mireya, Ira, Jacek
Unevictable pods recommendations in PerfectScale
Unevictable pods recommendations in PerfectScale