Operations | Monitoring | ITSM | DevOps | Cloud

Table of Contents A few months ago, we introduced Kubex support for the KAI Scheduler to improve GPU sharing for production inference workloads. The basic model is simple: The KAI Scheduler handles placement and GPU sharing. Kubex continuously observes usage and adjusts those allocations as demand changes. KAI provides the scheduling foundation. It lets multiple workloads share a GPU while accounting for the amount of GPU each workload requests. Kubex then closes the loop.

Table of Contents For most of Kubernetes’ history, memory management has been a blunt instrument. Cross your limit, and the kernel kills your container. There has been no equivalent to CPU throttling, no graceful backpressure, just a hard stop. With Kubernetes 1.37, that changes: Memory QoS, built on cgroups v2, graduates to Beta and is enabled by default.

This blog post is part of PagerDuty’s ongoing series on how we’re helping customers navigate their journey towards autonomous operations. Read on to learn about how Scoped OAuth for Public Apps, now in Early Access, builds towards this vision. Your security team asks a simple question during a routine review: which third-party apps can reach our PagerDuty data right now, and what exactly can they do with it?

This blog post is part of PagerDuty’s ongoing series on how we’re helping customers navigate their journey towards autonomous operations. Read on to learn about how PD Automation Runner, now in Early Access, builds towards this vision. Your on-call engineer gets paged for a critical alert on a self-hosted Kubernetes cluster. If this were a cloud-hosted service, Workflow Actions and the SRE Agent would already be working the problem: pulling logs, checking health, surfacing a fix.

The ability to log into all corporate applications with one username and password is pretty convenient, even compared to a password manager. As a benefit, the IT department can centrally enforce one desired two-factor auth mechanism. Now we, Icinga, also provide a so-called OpenID Connect integration for single sign-on. By the end of this text you’ll know how to connect your Icinga Web instance to the ID provider of your choice.

Most growing teams eventually consolidate onto a full observability platform, and for teams correlating logs, metrics, and traces across a complex system, that’s often the right call. But a dedicated log tool still wins for a specific set of teams: ones that need to move fast, keep costs simple, and get real answers from logs without carrying the weight of a platform they don’t fully need yet. Here’s when that’s you.