Modern systems
need modern tools
Modern systems
need modern tools
Modern systems
need modern tools
Honeycomb's datastore and query engine are purpose-built to give you fast answers to any question. Quickly understand performance and behavior down to a single user’s experience or by grouping together any combination of attributes. Now you can solve complex engineering mysteries you couldn’t solve before with traditional monitoring, metrics, and logging tools.
Compare: Analyzing Telemetry Data
Easily query and analyze complex telemetry data in seconds, with no limits on high-cardinality data or how many dimensions you group results by.
Capability
Honeycomb Approach
APM Approach
High-cardinality data analysis with high dimensionality
Group results by unlimited high-cardinality dimensions, query results still return in less than 3s
Slow, expensive analysis (if at all), costly dimensions, with pre-indexing required
Speed of query results
Fast, no matter how big your data set or how many high-cardinality dimensions you analyze
Results only return quickly when your data set is small
Speed of data query availability after ingestion
Newly ingested telemetry is available to query in under 5 seconds (near real-time)
Some telemetry available after seconds, some vendors need several minutes
Compare: Alerting for Issues in Production
Avoid on-call fatigue using debuggable alerts that matter the most to your users and warn you before service levels fall below acceptable targets.
Capability
Honeycomb Approach
APM Approach
Deciding what to monitor
Monitor what's important to users with service-level objectives (SLOs)
Monitor for known symptoms of past failures
Taking action on an alert
Debug SLO-triggered alerts using the same interface in the same tool
One tool triggers alerts, debugging using a different tool
Decreasing alert noise
Trigger single alerts before user experience falls under acceptable targets
Many error alerts are grouped together with AI deciding priority
Measuring failure or success
Uses individual requests to measure requests succeeded vs. requests failed
Uses metrics to measure good minutes vs. bad minutes
Compare: Isolating the Source of Issues
Any team member can use our consistent workflow for fast querying, heatmaps, and correlation detection to intuitively find the sources of novel problems.
Capability
Honeycomb Approach
APM Approach
Correlation detection
Multi-dimensional heatmaps, powered by machine analysis, let you decide what matters
Trust AI suggestions or single-dimension range selection
Triage workflow
Use one workflow, driven by querying and first principles, to surface the relevant telemetry data you need
Analyze different dashboards, then switch to other tools based on intuition
Know where to start
Quickly find the relevant data you need, regardless of prior experience
Relies on intuition and prior experience to determine which dashboards are most relevant
Finding telemetry data showing why an issue is occuring
Follow one workflow to surface relevant data and click directly into the telemetry you need
Manually check related dashboards, check logs, check traces, context-switch between many different tools
Compare: Organizational Adoption
New users can adopt Honeycomb without deep system familiarity, proprietary query languages, closed instrumentation, or any additional seat costs. We’re designed for easy adoption.
Capability
Honeycomb Approach
APM Approach
Supporting vendor-neutral standards
OpenTelemetry is the de facto standard for instrumentation with Honeycomb
Recommend proprietary agents and libraries, but may support OpenTelemetry
Cross-team adoption
Unlimited hosts, seats, users, etc. Simple event-based pricing encourages adoption
Price increases the more you use it (per host, seats, service, teams, etc.)
System familiarity required to triage
Anyone can triage correctly, regardless of familiarity
The less familiarity you have, the less likely you will triage correctly
Querying accessibility
Build queries intuitively with a visual UI
Learn proprietary query languages
Intended audience
Workflows for operations, development, product, support teams, and more
Focused on the needs of ops teams, with developer accessible tools like dashboards
Informing stakeholders
Technical teams and exec teams aligned on business goals with SLOs
Simplified exec dashboards to report current state as measured by metrics
Organizational memory
Learn by following past investigative steps with Team History; pin and save what’s useful
See and share dashboards with static views, lacking context for how they’re being used / what they mean
Compare: Analyzing Telemetry Data
Easily query and analyze complex telemetry data in seconds, with no limits on high-cardinality data or how many dimensions you group results by.
High-cardinality data analysis with high dimensionality
Group results by unlimited high-cardinality dimensions, query results still return in less than 3s
Speed of query results
Fast, no matter how big your data set or how many high-cardinality dimensions you analyze
Speed of data query availability after ingestion
Newly ingested telemetry is available to query in under 5 seconds (near real-time)
Compare: Alerting for Issues in Production
Avoid on-call fatigue using debuggable alerts that matter the most to your users and warn you before service levels fall below acceptable targets.
Deciding what to monitor
Monitor what's important to users with service-level objectives (SLOs)
Taking action on an alert
Debug SLO-triggered alerts using the same interface in the same tool
Decreasing alert noise
Trigger single alerts before user experience falls under acceptable targets
Measuring failure or success
Uses individual requests to measure requests succeeded vs. requests failed
Compare: Isolating the Source of Issues
Any team member can use our consistent workflow for fast querying, heatmaps, and correlation detection to intuitively find the sources of novel problems.
Correlation detection
Multi-dimensional heatmaps, powered by machine analysis, let you decide what matters
Triage workflow
Use one workflow, driven by querying and first principles, to surface the relevant telemetry data you need
Know where to start
Quickly find the relevant data you need, regardless of prior experience
Finding telemetry data showing why an issue is occuring
Follow one workflow to surface relevant data and click directly into the telemetry you need
Compare: Organizational Adoption
New users can adopt Honeycomb without deep system familiarity, proprietary query languages, closed instrumentation, or any additional seat costs. We’re designed for easy adoption.
Supporting vendor-neutral standards
OpenTelemetry is the de facto standard for instrumentation with Honeycomb
Cross-team adoption
Unlimited hosts, seats, users, etc. Simple event-based pricing encourages adoption
System familiarity required to triage
Anyone can triage correctly, regardless of familiarity
Querying accessibility
Build queries intuitively with a visual UI
Intended audience
Workflows for operations, development, product, support teams, and more
Informing stakeholders
Technical teams and exec teams aligned on business goals with SLOs
Organizational memory
Learn by following past investigative steps with Team History; pin and save what’s useful
Compare: Analyzing Telemetry Data
Easily query and analyze complex telemetry data in seconds, with no limits on high-cardinality data or how many dimensions you group results by.
High-cardinality data analysis with high dimensionality
Slow, expensive analysis (if at all), costly dimensions, with pre-indexing required
Speed of query results
Results only return quickly when your data set is small
Speed of data query availability after ingestion
Some telemetry available after seconds, some vendors need several minutes
Compare: Alerting for Issues in Production
Avoid on-call fatigue using debuggable alerts that matter the most to your users and warn you before service levels fall below acceptable targets.
Deciding what to monitor
Monitor for known symptoms of past failures
Taking action on an alert
One tool triggers alerts, debugging using a different tool
Decreasing alert noise
Many error alerts are grouped together with AI deciding priority
Measuring failure or success
Uses metrics to measure good minutes vs. bad minutes
Compare: Isolating the Source of Issues
Any team member can use our consistent workflow for fast querying, heatmaps, and correlation detection to intuitively find the sources of novel problems.
Correlation detection
Trust AI suggestions or single-dimension range selection
Triage workflow
Analyze different dashboards, then switch to other tools based on intuition
Know where to start
Relies on intuition and prior experience to determine which dashboards are most relevant
Finding telemetry data showing why an issue is occuring
Manually check related dashboards, check logs, check traces, context-switch between many different tools
Compare: Organizational Adoption
New users can adopt Honeycomb without deep system familiarity, proprietary query languages, closed instrumentation, or any additional seat costs. We’re designed for easy adoption.
Supporting vendor-neutral standards
Recommend proprietary agents and libraries, but may support OpenTelemetry
Cross-team adoption
Price increases the more you use it (per host, seats, service, teams, etc.)
System familiarity required to triage
The less familiarity you have, the less likely you will triage correctly
Querying accessibility
Learn proprietary query languages
Intended audience
Focused on the needs of ops teams, with developer accessible tools like dashboards
Informing stakeholders
Simplified exec dashboards to report current state as measured by metrics
Organizational memory
See and share dashboards with static views, lacking context for how they’re being used / what they mean
Don’t just take our word for it
Teams who switch to Honeycomb see significant increases in incremental revenue,
faster incident response, reduction in severe incidents, and decreased developer
burnout and churn. See what they have to say:
What would take me hours to debug before Honeycomb now takes me a couple of minutes, if that. The speed to get to the actual area where you’ve seen something, there’s just nothing like it.
There was no real way to find possible culprits with our classic APM. We had to know what we needed to find before we could find it—a dead end. With Honeycomb, our producers were delighted to see issues solved so quickly.
A lot of the data we care about most, like app IDs, have very high cardinality. We needed to drill down and group by data that helped us understand what individual customers are seeing happen with specific apps. Honeycomb’s ability to handle that turned out to be huge and incredibly useful for our future.
I keep thinking back to older problems, many took days or weeks to understand–we could have solved them in moments with Honeycomb.
Before we had Honeycomb, you used to have to know the schemas so you could optimize queries; you had to know what you needed to ask ahead of time.
Before Honeycomb, we would just speculate wildly about who was impacted by a given issue, or what changes would affect which customer. Now we know more, and guess less.
We’ve been told so many times that logging customer ID was impossible. Honeycomb just handles it. There have been outages where, without Honeycomb, it would have taken us significantly longer to get to the answer.
Honeycomb moved us in a direction of better and happier engineering outcomes. Honeycomb really fills a gap with my existing tools by providing insight into what’s actually going on inside applications. You look at the graphs and go, my application is doing what!? You can then go ahead and fix it.
Honeycomb saved us days of debugging.
What would take me hours to debug before Honeycomb now takes me a couple of minutes, if that. The speed to get to the actual area where you’ve seen something, there’s just nothing like it.
There was no real way to find possible culprits with our classic APM. We had to know what we needed to find before we could find it—a dead end. With Honeycomb, our producers were delighted to see issues solved so quickly.
A lot of the data we care about most, like app IDs, have very high cardinality. We needed to drill down and group by data that helped us understand what individual customers are seeing happen with specific apps. Honeycomb’s ability to handle that turned out to be huge and incredibly useful for our future.
I keep thinking back to older problems, many took days or weeks to understand–we could have solved them in moments with Honeycomb.
Before we had Honeycomb, you used to have to know the schemas so you could optimize queries; you had to know what you needed to ask ahead of time.
Before Honeycomb, we would just speculate wildly about who was impacted by a given issue, or what changes would affect which customer. Now we know more, and guess less.
We’ve been told so many times that logging customer ID was impossible. Honeycomb just handles it. There have been outages where, without Honeycomb, it would have taken us significantly longer to get to the answer.
Honeycomb moved us in a direction of better and happier engineering outcomes. Honeycomb really fills a gap with my existing tools by providing insight into what’s actually going on inside applications. You look at the graphs and go, my application is doing what!? You can then go ahead and fix it.
Honeycomb saved us days of debugging.
What would take me hours to debug before Honeycomb now takes me a couple of minutes, if that. The speed to get to the actual area where you’ve seen something, there’s just nothing like it.
There was no real way to find possible culprits with our classic APM. We had to know what we needed to find before we could find it—a dead end. With Honeycomb, our producers were delighted to see issues solved so quickly.
A lot of the data we care about most, like app IDs, have very high cardinality. We needed to drill down and group by data that helped us understand what individual customers are seeing happen with specific apps. Honeycomb’s ability to handle that turned out to be huge and incredibly useful for our future.
I keep thinking back to older problems, many took days or weeks to understand–we could have solved them in moments with Honeycomb.
Before we had Honeycomb, you used to have to know the schemas so you could optimize queries; you had to know what you needed to ask ahead of time.
Before Honeycomb, we would just speculate wildly about who was impacted by a given issue, or what changes would affect which customer. Now we know more, and guess less.
We’ve been told so many times that logging customer ID was impossible. Honeycomb just handles it. There have been outages where, without Honeycomb, it would have taken us significantly longer to get to the answer.
Honeycomb moved us in a direction of better and happier engineering outcomes. Honeycomb really fills a gap with my existing tools by providing insight into what’s actually going on inside applications. You look at the graphs and go, my application is doing what!? You can then go ahead and fix it.
Honeycomb saved us days of debugging.