Catch up with us, anytime and anywhere
29 August 2012
In the past year, the RCU lock-free hash table has been polished and made production-ready within the Userspace RCU project. It performs and scales really well for updates, key lookups and traversals in no particular key order, but does not fulfill ordered key traversal use-cases. This talk is presenting ongoing work on an ordered data structure that supports RCU reads: a cache-efficient, compact, fast, and scalable trie, inspired by Judy Arrays.
29 August 2012
The Linux Plumbers scaling micro-conference focus on scaling both upwards (many cores) and downwards (low footprint, energy efficiency) at all layers of the software stack. Our intent is to bring together application, libraries and kernel developers to discuss the scalability issues they currently face, and get exposure for the ongoing work on scalability infrastructure. It was held in August 29-31, in San Diego, California.
29 August 2012
LTTng 2.0 can be used with various tools which help digging through large amount of trace data, from high-level perspectives down to the details. This presentation will focus on the usability of LTTng, showing how the combined user-space and kernel tracers, high-level summary views such as LTTngTop, graphical analysis tools such as the Eclipse Linux Tools LTTng plugin, can be used to solve hard software problems.
30 April 2012
The newly available LTTng 2.0 kernel and user-space tracer offers a unified user interface tailored to the industry needs. It enables holistic tracing of live production systems across the kernel and user-space execution layers, while adding only a small overhead footprint. It is easy to install, uses most of the instrumentation sources present in mainline and distribution Linux kernels, and provides mechanisms for user-space instrumentation. This presentation covers the various features offered by LTTng 2.0. The target audience includes people responsible for maintenance, deployment or development of mission-critical, performance-sensitive, and/or latency-sensitive production systems running on GNU/Linux.
26 October 2011
This presentation introduces LTTng 2.0, detailing its new features. Amongst these, the most welcome are be the ability to use it on vanilla and distribution kernels, as well as the ability to hook on Tracepoints, Kprobes, Ftrace function tracing and Perf PMU counters. The new integrated command line interface for both the kernel and user-space tracers (LTTng and UST) will be presented. The new Common Trace Format (CTF) natively produced by LTTng 2.0 will be described. Conditional tracing within LTTng and UST will also be discussed. Tracing use-cases by distribution power-users, telecom, and open source projects (MariaDB, Qemu/KVM) will be presented. This talk will also cover trace analysis tools to handle many GB of trace data, trace data visualization such as statistics, event distribution, sequence diagram, correlation with GDB tracepoint and integration of text based logs with a text parser wizard. Target audience: anyone understanding performance issues on multi-core systems.