The document discusses enhanced Berkeley Packet Filter (eBPF) technology, which allows for advanced tracing and observability in Linux systems. It outlines the evolution of eBPF, its applications in performance analysis, network performance, security, and real-time monitoring, as well as the tools available for implementing it. Various command examples and visualizations for analyzing system behav

Application Pauses When Running JVM Inside Linux Control Groups Linux cgroups-based solutions (e.g., Docker, CoreOS) are increasingly being used to host multiple applications on the same host. We have been using cgroups at LinkedIn to build our own containerization product called LPS (LinkedIn Platform as a Service) and to investigate the impact of resource-limiting policies on application perform

What is NUMA NUMA NODES Cache Coherent NUMA (ccNUMA) Server Memory Hierarchy FileSystem Cache Latency Linux NUMA Policies Application Design NUMA Monitoring NUMA TESTING What is Numa In Symmetric Multiprocessor (SMP) systems all physical memory is seen as a single pool. All cpus share hardware resources and same address space – single instance of kernel. When physical memory and IO devices are equ

I must begin with saying that if you found this because you have a performance problem, you should almost certainly look elsewhere. It is highly unlikely that your performance problem is caused by your priority queue. If, however, you are curious, or you have done careful profiling and found out that the cache characteristics of your priority queue are causing your performance problem, and you can

You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session. Dismiss alert

Multi-core processors are capable of running multiple software streams/tasks concurrently. Multi-core allows a physical processor to simultaneously execute instructions from multiple processes or threads. Core of a processor is the part that executes application instructions. Core is shared by hardware threads (called Hyper-Threads). When two hyper-threads are active in the same core, it results i

Monitoring and Tuning the Linux Networking Stack: Receiving Data TL;DR This blog post explains how computers running the Linux kernel receive packets, as well as how to monitor and tune each component of the networking stack as packets flow from the network toward userland programs. UPDATE We’ve released the counterpart to this post: Monitoring and Tuning the Linux Networking Stack: Sending Data.

24 Aug 2015 on erlang garbage collection memory layout soft realtime Erlang Garbage Collection Details and Why It Matters One of the main problems that Erlang tried to solve was creating a platform for implementing Soft Realtime systems with a high level of responsiveness. Such systems require a fast Garbage Collection mechanism that doesn’t stop the system from responding in a timely manner. On o

09 Feb 2016 on erlang scheduling real-time preemptive migration Erlang Scheduler Details and Why It Matters There are some underlying features that make Erlang a soft real-time platform. One of them is its Garbage Collection mechanism which I talked about it in my previous article, Erlang Garbage Collection Details and Why It Matters. The other one is its scheduling mechanism that is well worth lo

To build for release cd release make make check You get two libraries in release/lib libsupermalloc.so requires transactional memory (haswell) libsupermalloc_pthread.so uses pthread locks I plan to add a runtime check and have only one library and an enviroment variable in the future. ------ You should also build and check with the internal testing code enabled. It's a different library partly bec