PerfectScale by doit logo
Platform
Pricing
Blog
PerfectScale by DoiT: Automated K8s Optimization and Management Platform

LIVE KUBERNETES WORKSHOP

In-place Pods Right-Sizing Without Restarts

Tuesday, September 30, 2025, 12:00 PM EST

 

 

Get Free Access

The release of in-place pod resizing in Kubernetes 1.33 introduces a powerful new capability: the ability to adjust container resource requests and limits without restarting pods. This breakthrough makes zero-downtime optimization finally possible, but it’s not a silver bullet. In-place resizing comes with caveats that, if misunderstood, can lead to instability or missed optimization opportunities.

That’s why PerfectScale’s approach has always centered on reliability-first optimization. While cost reduction is critical, it should never come at the expense of system performance or resilience. In a world where over-provisioning is the norm, often used as a safeguard against downtime, many organizations end up trading financial waste for a false sense of stability.

In this workshop, Anton Weiss and Arthur Berezin will unpack how to use in-place resizing effectively and, just as importantly, when not to use it. You’ll learn how PerfectScale’s implementation enhances this feature with intelligent safeguards, integrates into a proven 6-step optimization framework, and supports teams in making smarter, risk-aware decisions.

Whether you're running traditional workloads or exploring GPU-backed services, this session will help you align efficiency with reliability, not trade one for the other.

What We’ll Cover:

The Truth About In-Place Pod Resizing: What Kubernetes 1.33 enables, where in-place changes make a meaningful difference, and where they introduce risk.

The 6-Step Kubernetes Cost Optimization Framework: A proven, step-by-step methodology to drive efficiency without compromising service quality or SLOs.

GPU Workload Optimization Is Here (and Growing): With more workloads leveraging GPUs, visibility and right-sizing become even more critical—learn how to make smarter decisions in this emerging space.

Real-World Examples and Use Cases: See how engineering teams have applied these strategies to reduce costs while increasing system resilience and scalability.

Who Should Attend:

  • SRE/DevOps/Platform engineers looking for ways to improve their Kubernetes management practices.
  • FinOps practitioners wishing to understand Kubernetes cost components.
  • Data engineers wishing to leverage Kubernetes for AI/ML workloads.

Meet our experts

Arthur Berezin

Arthur Berezin, Product Leader & CNCF Ambassador

Arthur Berezin brings over 20 years of expertise in cloud and infrastructure technologies, having led transformative projects at JovianX, Cloudify, and Red Hat. As a seasoned entrepreneur and thought leader, he is passionate about leveraging his experience to drive innovation and deliver impactful products. Arthur's career is defined by his commitment to continuous learning and ability to navigate complex challenges. 

 

Anton Weiss

Anton Weiss, Chief Storyteller

Anton has a storied career in creating engaging and informative content that helps practitioners navigate through the complexities of ongoing Kubernetes operations. With previous experience as a CD Unit Leader, Head of DevOps, and CTO and CEO he has worn many hats as a consultant, instructor, and public speaker. He is passionate about leveraging his expertise to support the needs of DevOps, Platform Engineering, and Kubernetes communities.

reviews carousel

Want to Try PerfectScale for Free? 

Getting started takes less than 5 minutes. 

Simply install the PerfectScale Helm chart and watch the magic happens. 

No payment needed

FAQ

  1. What is PerfectScale?
    PerfectScale is an automated Kubernetes cost optimization and management platform. It's the industry's only production-ready automation platform that can safely and autonomously right-size your environment to enhance resilience and availability, eliminate waste, and reduce carbon emissions. With PerfectScale you can ensure your environment is always perfectly scaled to meet demand by effortlessly optimizing every layer of your K8s stack. 

  2. How do I implement PerfectScale?
    PerfectScale is a SaaS solution designed to simplify Kubernetes optimization. Getting started takes just a few minutes—deploy our lightweight agent to your target cluster, and optimization insights will immediately begin streaming to your organization's tenant.
     
  3.  We already use auto-scalers (HPA, KEDA, Karpenter), why do we need PerfectScale?
     While auto-scaling is important, proper configuration is complex. PerfectScale optimizes workloads, enhancing the efficiency of auto-scaling practices. Our recommendations respect your HPA settings to prevent conflicts, and work in tandem with Karpenter, right-sizing your requests automatically, and increasing savings.

  4.  How secure is PerfectScale?
    At PerfectScale, security and compliance are top priorities. Our SOC 2 Type 2 compliance ensures your data is protected. We also offer features like Single Sign-On (SSO), multi-factor authentication, and audit logs to help you use our platform securely.

  5. We are already using observability tools like Datadog, what's the added value? 
     While tools like Datadog or Prometheus are valuable, they often don’t provide actionable insights for continuous K8s optimization. PerfectScale offers recommendations and automated actions needed for optimal performance.

 

PerfectScale by DoiT, 2025