HPC
Profit and Loss Modeling on GPUs with ISO C++ Language Parallelism
The previous post How to Accelerate Quantitative Finance with ISO C++ Standard Parallelism demonstrated how to write a Black-Scholes simulation using ISO C++ standard parallelism with the code found in the /NVIDIA/accelerated-quant-finance GitHub repo. This approach enables you to productively write code that is both concise and portable. Using solely standard C++, it’s possible to write an application that can be run in parallel on a modern, multicore CPU or on a GPU without modification. This post builds a more complex model, starting from the previously developed parallel Black-Scholes code, and optimizes it to use the benefits of the GPU, while remaining in standard C++.
Read MoreHow to Accelerate Quantitative Finance with ISO C++ Standard Parallelism
Quantitative finance libraries are software packages that consist of mathematical, statistical, and, more recently, machine learning models designed for use in quantitative investment contexts. They contain a wide range of functionalities, often proprietary, to support the valuation, risk management, construction, and optimization of investment portfolios.
Read MoreSimplifying GPU Programming for HPC with NVIDIA Grace Hopper Superchip
The new hardware developments in NVIDIA Grace Hopper Superchip systems enable some dramatic changes to the way developers approach GPU programming. Most notably, the bidirectional, high-bandwidth, and cache-coherent connection between CPU and GPU memory means that the user can develop their application for both processors while using a single, unified address space.
Read MoreNew C++ Sender Library Enables Portable Asynchrony
NVIDIA is excited to announce the release of the stdexec library on GitHub and in the 22.11 release of the NVIDIA HPC Software Development Kit. The stdexec library is a proof-of-concept implementation of the Sender model (or Senders) for asynchronous computing—expected to become part of the C++ Standard in C++26. Senders help you separate the logic of your algorithm – the work you actually care about – from the details of where and how it gets executed. Unlike the C++17 parallel algorithms, with Senders you can chain multiple asynchronous computations (like GPU kernels) without unnecessary synchronization.
Read More