-
Mapping Plumbr to Monitoring Terminology
There are four “cornerstones” upon which the whole galaxy of monitoring services is built. These are Availability, Latency, Throughput, and Capacity. Whether our applications are
Filed under: Monitoring Performance Plumbr
-
Outage == Outrage
Let’s own up – we’ve all been part of teams that have faced failures and/or slow performance in web applications. The Google SRE book[1] highlights
Filed under: Monitoring Performance Plumbr Product Updates
-
Lies, damn lies and “our performance overhead is 2%”
Measuring the performance overhead of a Java agent happens to be a lot more complex exercise than it might originally seem. This post explains how system saturation is the key source for performance issues to be caused and gives specific examples how certain aspects introduced by agents can impact the performance of the system.
Filed under: Java Performance
-
The use of proxy indicators in service management
Proxy indicator is an indirect measure or sign that approximates or represents a phenomenon in the absence of a direct measure or sign. Lets look what this definition means in the context of service management and how we used proxy indicators ourselves in the absence of direct measure.
Filed under: Monitoring Performance
-
Going off-heap to improve latency and reduce AWS bill
Describing how going off-heap with the help of Chronicle Map helped us to get rid of the latency hiccups and cut our Amazon AWS bill in half.
Filed under: Java Performance
-
Most frequent performance bottlenecks
Explore the list of the performance issues with the most impact to end user experience. The data set originates from 1,020 different deployments Plumbr was monitoring during the past three months.
Filed under: Monitoring Performance
-
Squeezing data into the data structure
This story describes our recent experience with data structure optimization. With some tricks being pulled off the hat, we were able to reduce the original 2G data structure by 100x.
Filed under: Performance
-
Examples of caching backfiring on performance
In 2015 it should not surprise anyone that caching frequently used data is likely to improve the performance of the application. Caching certain data structures
Filed under: Monitoring Performance
-
Logging stop-the-world pauses in JVM
The points at which the application threads may be safely stopped are called safepoints. In this post we analyze the different reasons for the safepoints being triggered and show you a way to monitor your JVM for such events.
Filed under: Garbage Collection Monitoring Performance
-
How to think about performance requirements?
Performance requirements build the foundation for performance testing and production monitoring. The job of creating specific and measurable performance requirements is harder than it looks at the first place. This post will discuss important points that are often missed when performance requirements are set.
Filed under: Performance