Build Stuff'16 Lithuania

The Linux kernel has multiple "tracers" built-in, with various degrees of support for aggregation, dynamic probes, parameter processing, filtering, histograms, and other features. Starting from the venerable ftrace, introduced in kernel 2.6, all the way through eBPF, which is still under development, there are many options to choose from when you need to statically instrument your software with probes, or diagnose issues in the field using the system's dynamic probes. Modern tools include SystemTap, SysDig, ktap, perf, bcc, and others. In this talk, we will begin by reviewing the modern tracing landscape -- ftrace, perf_events, kprobes, uprobes, eBPF -- and what insight into system activity these tools can offer. Then, we will look at specific examples of using tracing tools for diagnostics: tracing a memory leak using low-overhead kmalloc/kfree instrumentation, diagnosing a CPU caching issue using perf stat, probing network and block I/O latency distributions under load, or merely snooping user activities by capturing terminal input and output.

When we talk about scaling, we usually mean up (bigger machine) or out (more machines). However, there is another alternative, which is changing our workload in a way that makes it inherently more parallelizable and then taking advantage of specialized hardware that's very good at handling that sort of thing. Most of us have exactly this type of hardware just sitting in our computers, doing very little most of the time. I'm talking, of course, about the GPU. General-purpose computing on the GPU (GPGPU) is no longer the domain of pure academic research. It is being used in real-world applications such as image processing and face recognition, cryptography, big data analysis, and Bitcoin mining. In this session we will examine the available GPGPU frameworks, learn how to integrate C++ AMP and OpenCL into regular .NET and Java applications, how to debug mixed .NET and GPGPU code in Visual Studio, and how to use the Aparapi framework to seamlessly mix regular CPU and GPGPU code in Java. We’ll even look at using GPGPU from server and client-side JavaScript with WebCL. We will discuss the kinds of tasks that can benefit from graphics card processing, how best to parallelize workloads, and what some of the performance trade-offs are.

Most of us are familiar with the architectural style known as REST, but even experienced developers often find it difficult to translate REST's architectural principles into running code. In this talk, we'll explore the elements of REST related to hypermedia and the principle of "hypermedia as the engine of application state" (HATEOAS) - we'll talk about why they matter, and when you might want to implement them in your own systems. We'll look at some of the tools that exist to help you design, deliver and debug your HTTP APIs, and we'll do some hands-on coding to show you how to implement lightweight, flexibile hypermedia patterns in your .NET web applications using the NancyFX HTTP framework and the HAL hypermedia application language.