Operations | Monitoring | ITSM | DevOps | Cloud

The latest Alloy mobile app updates are here, featuring push notifications in the Alloy Self-Service app and more to keep you productive on the go this summer!

Code in production will break. Maybe a request takes too long, maybe it fails quietly, or maybe it works fine one minute and falls over the next. Logs can help, sure—but they don’t always show the full picture, especially when performance issues are involved. Elastic APM gives you a clearer view. It traces what your application is doing from incoming requests to database queries and everything in between.

Running workloads on Google Kubernetes Engine (GKE) delivers impressive scalability and flexibility. Yet, it can also introduce a tricky challenge: tracking GKE costs accurately. Remember, GKE costs rarely scale linearly. Overprovisioned nodes, idle autoscalers, and orphaned workloads can quietly balloon your bill in the background. And while GKE’s native tools offer some visibility, they often miss the full picture.

Sovereign AI ensures control over AI/ML data, models, and infrastructure, which is now essential for enterprises, regulated industries, and national interests. JFrog and NVIDIA have collaborated to deliver a secure, scalable solution for sovereign AI. NVIDIA provides the accelerated computing and AI software while JFrog ensures trusted DevSecOps and MLOps practices across the entire AI lifecycle, from model development and security scanning to deployment at the edge and in air-gapped environments.

OpenSearch is a powerful, scalable search and analytics engine that can do amazing things for logging, observability, and full-text search. But like any distributed system, it only performs well if you keep it properly tuned and healthy. Ignore it, and you risk slower queries, higher costs, and even data loss. Here are five practical tips to keep your OpenSearch cluster running smoothly and efficiently.

In modern software development, different environments often have different compliance requirements. Your development environment might allow more flexibility, while production demands strict controls around security scans, testing, and code review. Environment Policy helps you codify these requirements and enforce them consistently.

Across the Asia Pacific region significant investment is going into new subsea cable infrastructure that will help sustain the long-term demand for AI. We’ve written a lot on this blog about the impact of AI on networks and how AI workloads require low latency, high-capacity data transfer. This in turn puts more pressure on existing network infrastructure and in particular subsea cable systems - which provide the global backbone for cloud platforms and data centres.

Infrastructure management isn't about taking sides. Too often, teams get pulled into “X is better than Y” debates that miss the bigger picture: your compute stack should serve your needs, not industry hype. A common decision point in the past has been the choice between bare metal or cloud hyperscalar virtualization. Nowadays, the answer isn't 1 or 0.

If you’re working with metrics in Grafana Cloud, chances are you’ve come across DPM (data points per minute). It shows up in usage dashboards, invoice breakdowns, and occasionally pops up in Slack when your ingestion numbers start looking suspicious. DPM can also be seen in the Grafana Cloud billing and usage dashboard, which is available by default in every Grafana Cloud account. It helps you understand how much data you’re sending—and whether it’s more than you need.

Private cloud infrastructure adoption is accelerating rapidly. This move is driven by the ongoing “cloud reset” as leaders rethink their hybrid and multi-cloud strategies, seeking greater control, security, and flexibility for their IT workloads. As a matter of fact, leaders in 69% of organizations are considering repatriating workloads, and one-third already have.