We run a lot of microservices. Image processing, AI inference, image delivery and API serving - all separate, and keeping an eye on all of them was getting complicated. CubeAPM helped bring that together. Traces, logs and infra in one place, and the correlation between them is what actually helps when something in the pipeline breaks.

Support is the other thing. We have a shared Slack with their team and I'd say the response time is genuinely fast. Not the usual open-a-ticket-and-wait experience.

And the migration was smoother than I expected. We were already on OpenTelemetry so most of our instrumentation just carried over. Review collected by and hosted on G2.com.

GPU monitoring. We run inference workloads on GPU machines and I can't see GPU utilization inside CubeAPM, so I still have to check that elsewhere. Hoping they add this at some point. Review collected by and hosted on G2.com.

CubeAPM provides full-stack observability in a single platform, including traces, metrics, logs, infrastructure monitoring, and real-user monitoring, which helps us avoid using multiple monitoring tools. It was easy to implement and integrate with our services, and the interface makes it simple to investigate issues across logs, traces, and metrics. Running CubeAPM in our own cloud is also valuable for security and compliance. We use it daily for monitoring and debugging, and the support team has been responsive and helpful whenever needed. Review collected by and hosted on G2.com.

The core observability features are strong overall, but there are still a few areas that could be improved. Real User Monitoring (RUM) and synthetic monitoring, in particular, feel fairly basic right now, and adding more capabilities in these areas would make the platform even more powerful. Review collected by and hosted on G2.com.

CubeAPM provides a great observability platform that makes it very easy to monitor application performance in real time. Since it can run within an entity's own cloud, is huge plus as data stays local and compliance is easier. Onboarding and integration was compatible with open telemetry (supports Datadog/relic/prometheus agents) so we didnt have remove existing instrumentation . UI/Dashboards are clean. Excellent customer support / handholiding and response times are in minutes. Review collected by and hosted on G2.com.

Cloud cost visibility / recommendations to optimise Review collected by and hosted on G2.com.

CubeAPM gives us strong observability for APM along with application runtime metrics within a single platform. We now have clear visibility into system performance and bottlenecks, which has significantly reduced downtime for our critical services. The support team is responsive, often available on Slack, and quick to assist whenever issues arise. CubeAPM’s predictable pricing and OTEL-native integrations made adoption smooth and kept costs under control compared with other legacy tools. Review collected by and hosted on G2.com.

The UI and dashboard experience could be more polished and intuitive. A bit more consistency in layout and interactions would improve usability. Additionally, AI-driven observability features (such as automated root-cause analysis or smarter anomaly detection) could be further strengthened to add more proactive insights. Review collected by and hosted on G2.com.

CubeAPM gives us full-stack observability across traces, metrics, logs and infrastructure - all under one roof. Don't have to juggle separate tools for different purposes. For a lean team like ours building an AI language learning app, having dashboards that actually help you spot latency spikes or error patterns quickly is important.

The fact that it runs inside our own cloud is a big plus. Our users' voice recordings and learning data stays in our own infrastructure. And since CubeAPM is natively compatible with OpenTelemetry, we didn't have to remove or redo our existing instrumentation when we migrated. That saved us a lot of time.

Support has been excellent. Previously we had to raise a ticket and wait. With CubeAPM the response is in minutes on a shared Slack channel with their engineering team. Review collected by and hosted on G2.com.

For dashboarding I want to have more visualization options. Basic ones are there but advanced ones like pie charts and histograms are not available yet. Not a blocker for us but would be good to have. Review collected by and hosted on G2.com.

CubeAPM provides excellent support, which is rare to find from other providers unless your account is very large. We have dedicated Slack channels where engineers from both teams are present, and whenever we reach out with an issue, it is resolved within a reasonable timeframe. The implementation process was also quick, as CubeAPM is compatible with our existing New Relic agents, which saved us from having to redo a lot of work on our end. Review collected by and hosted on G2.com.

There are two main points I want to mention. First, their AI-based functionalities still seem somewhat immature. Second, they do not offer SIEM monitoring. However, for our needs, these issues are not dealbreakers. Review collected by and hosted on G2.com.

What I like most about CubeAPM is that it’s cost-effective and significantly cheaper than some other tools, where the billing can be difficult to understand and justify. It’s also self-hosted, which means we stay in full control of our data. Support has been solid as well: the CubeAPM team and our team communicate through a Slack channel, and they respond quickly whenever we run into issues. Review collected by and hosted on G2.com.

CubeAPM has a good user experience overall, but the UI could be improved. It really needs a designer’s touch to feel more polished and consistent. Review collected by and hosted on G2.com.

CubeAPM brings logs, metrics, and traces together in one place, which makes it much easier to understand what’s happening across the system. The way different data points are correlated helps speed up troubleshooting and improves overall visibility. Review collected by and hosted on G2.com.

One area where it could improve is AI-driven observability. Since those capabilities are limited, we still need to rely on an additional tool for more advanced AI-based analysis. Review collected by and hosted on G2.com.

CubeAPM is a unified tool to get all round monitoring of applications, logs, infrastructure and mobile apps. It gives me a unified view in a single dashboard which makes RCA easy. The support is really fast. We use a common teams channel. Review collected by and hosted on G2.com.

CubeAPM doesn’t provide any tools around LLM observability. We use LLMs like ChatGPT and Gemini and need to keep track of the token usage. We need to depend on another tool to monitor our LLM calls. Review collected by and hosted on G2.com.

One tool for metrics, logs, and traces—without the Datadog bill. Review collected by and hosted on G2.com.

Honestly, the support is what stands out most. We’ve had shared Slack channels with other APM vendors, and most of them go quiet after the first week. With CubeAPM, the team is actually there—when we reach out with an issue or question, someone responds within minutes. That’s not something I expected going in.

The observability coverage is solid, too. We run several microservices—image processing, format conversion, video transcoding, and our DAM APIs. Before CubeAPM, keeping tabs on all of them meant constantly switching between tools. Now traces, logs, and metrics are all in one place, and the correlation between a latency spike in the processing API and downstream errors becomes obvious pretty quickly. It makes RCA a lot faster.

The migration was also smoother than I thought it would be. We were already using OpenTelemetry across most of our services, so we didn’t have to rip out existing instrumentation. We just pointed it at CubeAPM and got going. Review collected by and hosted on G2.com.