We are currently #profiling #Snack with #JProfiler. Love that #jdbc #hotspot #view.
#health #diet #business #marketing #product #system #knowledge #information #data #linearprogramming #optimization #price #cost #ai #artificialintelligence #superintelligence
#health #diet #business #marketing #product #system #knowledge #information #data #linearprogramming #optimization #price #cost #ai #artificialintelligence #superintelligence
November 5, 2026 at 2:08 PM
We are currently #profiling #Snack with #JProfiler. Love that #jdbc #hotspot #view.
#health #diet #business #marketing #product #system #knowledge #information #data #linearprogramming #optimization #price #cost #ai #artificialintelligence #superintelligence
#health #diet #business #marketing #product #system #knowledge #information #data #linearprogramming #optimization #price #cost #ai #artificialintelligence #superintelligence
This workflow became possible in JProfiler 16.3, which makes all major JProfiler subsystems available to AI agents. In JProfiler 16.1, we released the first version of our MCP server. It focused on the agent workflow with a limited set of data.
November 6, 2026 at 2:44 PM
This workflow became possible in JProfiler 16.3, which makes all major JProfiler subsystems available to AI agents. In JProfiler 16.1, we released the first version of our MCP server. It focused on the agent workflow with a limited set of data.
Read the full story:
www.ej-technologies.com/blog/2026/1...
Set up JProfiler for your agent with a single command:
www.ej-technologies.com/jprofiler/a...
www.ej-technologies.com/blog/2026/1...
Set up JProfiler for your agent with a single command:
www.ej-technologies.com/jprofiler/a...
JProfiler Makes Heap Dumps Useful to AI Agents
JProfiler 16.3 makes all major JProfiler subsystems available to AI agents through its MCP server. One of the most useful additions is heap dump analysis. Agents can now inspect the objects in a heap, follow their references and find out why they were not garbage collected.Without runtime data, an AI agent cannot see which objects are alive or what keeps them reachable. JProfiler closes that gap, so the agent can start with actual evidence and use it to find the relevant source code.
www.ej-technologies.com
November 6, 2026 at 2:44 PM
Read the full story:
www.ej-technologies.com/blog/2026/1...
Set up JProfiler for your agent with a single command:
www.ej-technologies.com/jprofiler/a...
www.ej-technologies.com/blog/2026/1...
Set up JProfiler for your agent with a single command:
www.ej-technologies.com/jprofiler/a...
In JProfiler 16.3, we adopted a new approach: a single query tool now provides access to a broad range of datasets. Each query returns a JSON document. The agent uses jq, a query language for JSON, to filter, sort and limit the data inside JProfiler.
November 6, 2026 at 2:44 PM
In JProfiler 16.3, we adopted a new approach: a single query tool now provides access to a broad range of datasets. Each query returns a JSON document. The agent uses jq, a query language for JSON, to filter, sort and limit the data inside JProfiler.
midden - a fast JVM heap dump analyzer
Written in Rust, only takes fraction of the times of MAT, JProfiler or VisualVM. Guess I'll test this quite soon.
github.com/NullByte3/mi...
Written in Rust, only takes fraction of the times of MAT, JProfiler or VisualVM. Guess I'll test this quite soon.
github.com/NullByte3/mi...
GitHub - NullByte3/midden: Fast JVM heap dump analyser: retained sizes, owners and leak suspects from .hprof files. One static binary, no JDK needed.
Fast JVM heap dump analyser: retained sizes, owners and leak suspects from .hprof files. One static binary, no JDK needed. - NullByte3/midden
github.com
November 6, 2026 at 1:56 PM
midden - a fast JVM heap dump analyzer
Written in Rust, only takes fraction of the times of MAT, JProfiler or VisualVM. Guess I'll test this quite soon.
github.com/NullByte3/mi...
Written in Rust, only takes fraction of the times of MAT, JProfiler or VisualVM. Guess I'll test this quite soon.
github.com/NullByte3/mi...
This also opened up heap dump analysis to agents. JProfiler-specific jq functions let the agent follow paths to GC roots, aggregate incoming references, inspect retained object graphs and retrieve inspection results.
November 6, 2026 at 2:44 PM
This also opened up heap dump analysis to agents. JProfiler-specific jq functions let the agent follow paths to GC roots, aggregate incoming references, inspect retained object graphs and retrieve inspection results.
An AI agent recently used JProfiler 16.3 to fix a memory leak in our own install4j runtime by analyzing an HPROF heap dump from a real support case.
November 6, 2026 at 2:44 PM
An AI agent recently used JProfiler 16.3 to fix a memory leak in our own install4j runtime by analyzing an HPROF heap dump from a real support case.
"Agents Have No Runtime Insight"-an airhacks.fm podcast conversation with Ingo Kegel is ready to listen: 👉 adambien.blog/roller/agent...
#java #podcast #jprofiler #airhacks
#java #podcast #jprofiler #airhacks
airhacks.fm podcast
podcast with adam bien
airhacks.fm
October 18, 2026 at 10:51 AM
"Agents Have No Runtime Insight"-an airhacks.fm podcast conversation with Ingo Kegel is ready to listen: 👉 adambien.blog/roller/agent...
#java #podcast #jprofiler #airhacks
#java #podcast #jprofiler #airhacks
jvmguard 0.3 is out with OpenTelemetry support. Methods annotated with "WithSpan" or "Observed" now automatically become transactions in jvmguard, and you can use them to trigger deep profiling captures (heap dumps, JFR snapshots, JProfiler snapshots) for anomalous situations.
October 13, 2026 at 1:41 PM
jvmguard 0.3 is out with OpenTelemetry support. Methods annotated with "WithSpan" or "Observed" now automatically become transactions in jvmguard, and you can use them to trigger deep profiling captures (heap dumps, JFR snapshots, JProfiler snapshots) for anomalous situations.
However, times did not come down a lot. The next profiling run with JProfiler showed that the base64 decoder was exceptionally slow, taking a whopping 85% of the entire run.
September 7, 2026 at 10:05 AM
However, times did not come down a lot. The next profiling run with JProfiler showed that the base64 decoder was exceptionally slow, taking a whopping 85% of the entire run.
Time to get serious: process explorer from sysinternals should show which thread is consuming the cpu cycles. If it’s indeed adt ie a Java process, then hardcore profilers dig up details (like Jprofiler). Then what? Seems like a waste of time. Faster go to Costco and buy a new machine
January 11, 2025 at 10:56 PM
Time to get serious: process explorer from sysinternals should show which thread is consuming the cpu cycles. If it’s indeed adt ie a Java process, then hardcore profilers dig up details (like Jprofiler). Then what? Seems like a waste of time. Faster go to Costco and buy a new machine
First time using JProfiler. Was very impressed at how easy it was to connect to my Ktor service running in k8s.
Now i just have to track down my memory issue...
Now i just have to track down my memory issue...
November 11, 2024 at 4:56 AM
First time using JProfiler. Was very impressed at how easy it was to connect to my Ktor service running in k8s.
Now i just have to track down my memory issue...
Now i just have to track down my memory issue...
#yourkit vs #jprofiler is the question...
November 22, 2024 at 11:13 PM
#yourkit vs #jprofiler is the question...
##Programacion ##BuenasPracticas Realiza profiling y benchmarking de código crítico para identificar cuellos de botella y validar mejoras de rendimiento, usando herramientas como JProfiler, gprof, cProfile o EXPLAIN en SQL.
April 21, 2026 at 12:07 PM
##Programacion ##BuenasPracticas Realiza profiling y benchmarking de código crítico para identificar cuellos de botella y validar mejoras de rendimiento, usando herramientas como JProfiler, gprof, cProfile o EXPLAIN en SQL.
"JProfiler Visual Studio Code Integration -- The Kotlin Multiplatform Killer Use Case" -- an airhacks.fm podcast conversation with Ingo Kegel / @ingokegel.bsky.social
is ready to listen 👉 adambien.blog/roller/jprof... #java #podcast #airhacks #kotlin
is ready to listen 👉 adambien.blog/roller/jprof... #java #podcast #airhacks #kotlin
airhacks.fm podcast
podcast with adam bien
airhacks.fm
September 11, 2025 at 4:47 PM
"JProfiler Visual Studio Code Integration -- The Kotlin Multiplatform Killer Use Case" -- an airhacks.fm podcast conversation with Ingo Kegel / @ingokegel.bsky.social
is ready to listen 👉 adambien.blog/roller/jprof... #java #podcast #airhacks #kotlin
is ready to listen 👉 adambien.blog/roller/jprof... #java #podcast #airhacks #kotlin
Get 50% off with the coupon code: "java2023" for JProfiler 👉 www.ej-technologies.com Thanks to @ingokegel.bsky.social for the offer.
November 19, 2023 at 4:44 PM
Get 50% off with the coupon code: "java2023" for JProfiler 👉 www.ej-technologies.com Thanks to @ingokegel.bsky.social for the offer.
"How JProfiler Happened" 👉 airhacks.fm#episode_265 Podcast with @ingokegel.bsky.social #java #podcast
October 31, 2023 at 2:19 PM
"How JProfiler Happened" 👉 airhacks.fm#episode_265 Podcast with @ingokegel.bsky.social #java #podcast
How a 1-pizza team created a great profiler: "How JProfiler happened" 👉 an airhacks.fm conversation with Ingo Kegel is ready to listen: adambien.blog/roller/abien... #java #podcast
airhacks.fm podcast
podcast with adam bien
airhacks.fm
October 22, 2023 at 1:01 PM
How a 1-pizza team created a great profiler: "How JProfiler happened" 👉 an airhacks.fm conversation with Ingo Kegel is ready to listen: adambien.blog/roller/abien... #java #podcast
⏺️ JFR recordings, HPROF heap dumps and JProfiler snapshots can be triggered by humans in the jvmguard UI, by coding agents through the MCP server, or automatically when a trigger fires.
August 7, 2026 at 3:21 PM
⏺️ JFR recordings, HPROF heap dumps and JProfiler snapshots can be triggered by humans in the jvmguard UI, by coding agents through the MCP server, or automatically when a trigger fires.
I have spent more than two decades developing a profiler for the JVM and a lot of folks here will know JProfiler: a development profiler that provides the kind of deep insights that you need to solve hard problems.
August 7, 2026 at 3:21 PM
I have spent more than two decades developing a profiler for the JVM and a lot of folks here will know JProfiler: a development profiler that provides the kind of deep insights that you need to solve hard problems.
You know, this is really intriguing and I will see if we can come up with something for JProfiler 13.3. Thanks for the inspiration.
August 5, 2026 at 10:49 AM
You know, this is really intriguing and I will see if we can come up with something for JProfiler 13.3. Thanks for the inspiration.
Exactly right, and this is why container memory limits trip up so many otherwise careful setups. We run into this constantly when profiling JVMs in Kubernetes clusters, in case this is useful: www.ej-technologies.com/blog/2022/12...
Profiling Java applications in a Kubernetes cluster
This screencast shows how you can profile JVMs running in Kubernetes cluster with JProfiler. A profiling session with a note taking demo application sessions is started from the IDE, which provides ad...
www.ej-technologies.com
August 4, 2026 at 6:44 PM
Exactly right, and this is why container memory limits trip up so many otherwise careful setups. We run into this constantly when profiling JVMs in Kubernetes clusters, in case this is useful: www.ej-technologies.com/blog/2022/12...
To read the full story with all numbers, see my blog post at www.ej-technologies.com/blog/2026/0...
Making the JProfiler MCP Server Robust for Weaker Models
Using flagship models is not ideal for all coding tasks. They are slow and they cost a lot. So many tasks, especially chores, are more suitable for lower-tier models that are more nimble and do the task at a fraction of the cost. However, the tool-calling behavior of these models can be more erratic than what the top models are capable of.In JProfiler 16.1 we added an MCP server to let AI coding agents profile Java applications and analyze the results. This has been met with a lot of positive feedback from our customers, so it is an area that we intend to expand on in the upcoming releases.For the initial release, most of our testing had been done mainly with Claude Code. For JProfiler 16.2, we have made the MCP server a lot more robust for other AI models. Read on for what the other models did wrong and how we made the profiling workflow robust against their shortcomings.
www.ej-technologies.com
August 3, 2026 at 3:23 PM
To read the full story with all numbers, see my blog post at www.ej-technologies.com/blog/2026/0...
Failures of these kinds are expensive. A failing GPT run needed 42 tool calls and 4 minutes on average, about eight times the cost of a passing run. The whole GPT test matrix cost $13.21 with JProfiler 16.1.1 and $3.13 with JProfiler 16.2.
August 3, 2026 at 3:23 PM
Failures of these kinds are expensive. A failing GPT run needed 42 tool calls and 4 minutes on average, about eight times the cost of a passing run. The whole GPT test matrix cost $13.21 with JProfiler 16.1.1 and $3.13 with JProfiler 16.2.
In JProfiler 16.2, we hardened this workflow. To show the improvements, we ran four models against the 16.1.1 and the 16.2 MCP server. GPT went from 38/55 to 55/55, DeepSeek and Qwen improved as well, while Kimi was reliable throughout.
August 3, 2026 at 3:23 PM
In JProfiler 16.2, we hardened this workflow. To show the improvements, we ran four models against the 16.1.1 and the 16.2 MCP server. GPT went from 38/55 to 55/55, DeepSeek and Qwen improved as well, while Kimi was reliable throughout.