Quickly explain latency, traffic anomalies, error rates, or availability issues
BubbleUp dramatically speeds up debugging by automatically detecting commonalities in outliers. Highlight anomalies on any heatmap visualization or query result, and BubbleUp compares thousands of high-cardinality and high-dimensionality events inside and outside the selection, surfacing the highest differences for you because those are most likely to explain the issue.
Honeycomb’s BubbleUp gives you confidence that you found the problem and that you’re smart enough to find more. You solve one thing that seemed intractable.
You realize you’re smarter than you thought you were with this new tool. The new confidence when building things is wonderful.
Ryan K., Senior Staff Software Engineer
Anyone can investigate issues faster, regardless of familiarity with code
Honeycomb overcomes the bottleneck of using your most tenured engineers to investigate issues. Other tools limit querying by cardinality, dimensionality, or speed so you are forced to rely on an expert to point your investigation in the right area.
BubbleUp shatters that bottleneck. Anyone, regardless of familiarity with your code, can investigate any new issue across all your data from a few initial clues. Draw a box around what looks strange in a heatmap, or select a group that stands out from a query result, and BubbleUp cycles through all possible dimensions to surface common patterns and quickly locate the source of issues in your system. There is no need to rely on intuition and prior experience. Anyone can use the same methodology, every time. In practice, this is an incredibly fast and effective way of finding answers to any possible question.
BubbleUp solves what most
AIOps solutions get wrong
Honeycomb believes in a performant symbiosis of machine intelligence combined with human intelligence.
Other solutions rely on historical data to automatically detect and resolve future issues, which might work if systems behaved consistently over time. In modern architectures, however, system behavior changes frequently so an algorithm will think every new thing is an anomaly.
What is missing is today’s context. Computers excel at churning through large datasets looking for patterns. Humans excel at putting those patterns into context and finding the ones that matter.
BubbleUp combines both. Select suspicious performance bottlenecks on a heatmap, expose the common patterns in their data, and decide what is important. Then triage, identify, and resolve. Blazingly fast.