https://schedule.cppnow.org/session/2026/benchmarking/

Every low-latency C++ developer measures time – but how many truly understand what they’re measuring?

This talk takes a deep dive into the machinery behind performance measurement on modern Linux systems: from std::chrono::steady_clock::now() through the vdso, down to rdtsc and hardware performance counters, and back up through the layers of abstraction that can silently distort your results.

Along the way, we’ll explore the subtle pitfalls that make microbenchmarking so treacherous: compilers that optimise away the code you’re trying to measure, branch predictors that learn your benchmark’s patterns, pipelining effects that make sequential measurements misleading, and the surprisingly complex question of what “now” even means on a modern CPU.

Matt Godbolt

Matt Godbolt is the creator and primary maintainer of Compiler Explorer. A C++ developer based in Chicago, he works at Hudson River Trading and is one half of the Two’s Complement podcast. His career has been a steady journey deeper into the machine: from writing rendering engines for video games to years of low-latency trading systems. His recent Advent of Compiler Optimisations series explored things compilers do that feel like magic.

When not hacking on Compiler Explorer, Matt likes to write emulators for the computers he grew up on, the ZX Spectrum, BBC Micro and Sega Master System.

View The Full Schedule Here

Join Us in Aspen, Colorado USA for C++Now 2026. 45+ Confirmed C++ Sessions, 3 Keynotes & Lightning Talks Await You!

https://schedule.cppnow.org/session/2026/making-cpp-standard-parallelism-multidimensional/

Standard C++ currently exposes potential parallelism in a few different ways: explicit thread creation, the parallel algorithms, std::execution, and std::simd. We can think of natural ways to extend and improve each of these. Such efforts are worth continuing, but what is missing here is multidimensionality. This effectively excludes a wide class of algorithms that have been proven optimal in terms of parallel synchronization and memory movement. We suggest three topic areas for incorporating multidimensionality into the Standard’s approaches to parallelism.

Locality: Constructing optimal multidimensional algorithms calls for exploiting locality multidimensionally, analogous to how std::simd does this for one dimension

Disjointness: Multidimensional algorithms can expose parallelism while helping users and implementations avoid complicated reasoning about whether possibly multidimensional arrays “overlap”

Coupling of low-level parallelism and memory operations: this leads to a preference for hiding layouts of local multidimensional arrays, and a proliferation of compile-time parameters that calls for just-in-time compilation

We will explain how Standard C++ already has the seeds of these three properties, and suggest ways to extend the Standard so that users can write portable code that performs well on a variety of hardware.

Mark Hoemmen

Mark Hoemmen is a programming models architect with a background in parallel computing and numerical linear algebra. He joined NVIDIA in 2022, and his current role there is to bridge between the C++ Standard and his company’s parallel programming models.

Mark finished his PhD on “communication-avoiding” linear algebra algorithms in 2010. After that, he worked for ten years for Sandia National Laboratories, where he did research on communication-avoiding and fault-tolerant algorithms, and contributed to several internal and open-source scientific computing software projects. He then took a position at Stellar Science (a small Albuquerque company) for two years, focusing on C++ modernization and performance measurement, before moving to NVIDIA.

Mark’s preferred programming language is C++. He has been writing it professionally for over 25 years, and has been contributing to the C++ Standard (WG21) process since 2017. Mark is main author of std::linalg, a linear algebra library that was added to C++26. He is also coauthor of several proposals relating to multidimensional arrays (such as mdspan (in C++23) and submdspan (in C++26)) and parallel algorithms.

View The Full Schedule Here

Join Us in Aspen, Colorado USA for C++Now 2026. 45+ Confirmed C++ Sessions, 3 Keynotes & Lightning Talks Await You!

https://schedule.cppnow.org/session/2026/reflection-is-only-half-the-story/

C++26 gives us reflection, the culmination of decades of work building up support for more and more compile-time programming in C++. But reflection primarily only lets us *observe*. The important next question is: what might it look like if we were to *generate*?

This talk is not a pitch for a specific direction for any particular future C++ standard, but rather a tour of the design space of source code generation for C++. We have some options for code generation today — macros and templates. Where do they hit their limits? What can we learn from our peers in the programming language world — like Rust, Swift, and D? Is there any low-hanging fruit to be found?

There are many facets to consider: capability, composability, cohesion, debuggability, ergonomics, error quality, teachability, tooling. The goal of this talk is to leave with better instincts — and better questions — for evaluating source code generation for C++ as reflection moves from novelty to foundational.

Barry Revzin
Barry is a senior C++ developer at Jump Trading in Chicago, a technology and engineering-driven trading firm. After programming for many years, he got really into the nuances and intricacies of C++ by being unreasonably active on StackOverflow (where he is the top contributor in every standard from C++14 up through C++26). A lot of his C++ knowledge comes from just answering questions that he doesn’t know the answers to, especially when he answers them incorrectly at first.

His C++ involvement escalated when he started attending standards committee meetings in 2016, having written dozens of papers for C++20, C++23, and now C++26. You might know him from such features as deducing this, explicit(bool), conditionally trivial special member functions, if consteval, packs in structured bindings, and now, reflection.

Outside of the C++ world, Barry is an obsessive swimming fan. He writes fun data articles for SwimSwam, which led him to be able to do work with a briefly-existing professional swim team (the DC Trident). He also periodically gets to interview a bunch of Olympians.

View The Full Schedule Here

Join Us in Aspen, Colorado USA for C++Now 2026. 45+ Confirmed C++ Sessions, 3 Keynotes & Lightning Talks Await You!

These Are Just Some Of The Accepted Sessions For C++Now 2026!

View Confirmed Sessions

This year we will have 3 tracks with 48 outstanding presentations from C++ experts, including three keynotes, as well as lightning talks in the evening, a “Future of C++” panel, and the “Project in a Week” hands-on workshop!

If you’re still on the fence about whether to join us, now is a good time to choose to attend!  Check out the attendee video.  The venue has a hard attendee limit of 140 this year, and if the conference sells out (as it tends to do), your remaining option is to wait until 2027! Secure your place now at this amazing C++ Conference, held at a one of a kind venue in the beautiful and iconic landscape of the Rocky Mountains. There is no other event quite like it!

Held May 4–8, 2026, C++Now brings together library authors, compiler developers, and expert C++ practitioners for five days of deep technical content, collaboration, and discussion in a uniquely intimate setting.

The conference takes place at the Aspen Center for Physics in Aspen, Colorado, providing an inspiring environment for learning, networking, and pushing the boundaries of modern C++.

Highlights include:

• In-depth technical sessions and tutorials
• Lightning talks and interactive discussions
• The “Future of C++” panel
• Collaborative events like Project in a Week
• Social events and networking with leading C++ experts

Join us in Aspen and be part of the conversation shaping the future of C++.

Register now for $1999 including T-shirt, Hoodie + Early Access to the C++Now 2026 YouTube Videos (Does not include Lodging)