GPU Tile 1 Tile 0 Xe LinkProject Goals - Offer high-level, standard C++ distributed data structures - Support distributed algorithms - Achieve high performance for both multi-GPU, NUMA, and multi-node reduce(par_unseq, z, 0, std::plus()); }Outline - Background (Ranges, Parallelism, Distributed Data Structures) - Distributed Ranges (Concepts) - Implementation (Algorithms and views) - Complex sparse matrices) - Lessons learnedOutline - Background (Ranges, Parallelism, Distributed Data Structures) - Distributed Ranges (Concepts) - Implementation (Algorithms and views) - Complex

C++20 ❤ SQL John R Bandela, MDOverview u SQL u C++ Example u Implementation techniques u fixed_string u meta_struct u Parsing compile time strings into meta_structsOverview v SQL v C++ Example “CppCon”Query for orders with input from user SELECT orders.id, name, item, price, discount_code FROM orders JOIN customers ON customers.id = customerid WHERE price > {price_from_user};Avoiding SQL Injection item, price, discount_code FROM orders JOIN customers ON customers.id = customerid WHERE price > ?;SQL Library Options u Traditional database library with strings u Domain specific language u Object

program for a supercomputer? Introduce PGAS Model, RDMA Building Remote Pointer Types Building Distributed Data Structures Extending to GPUsThis Talk Background: how do we write a program for a supercomputer supercomputer? Introduce PGAS Model, RDMA Building Remote Pointer Types Building Distributed Data Structures Extending to GPUsThis Talk Background: how do we write a program for a supercomputer? Introduce Introduce PGAS Model, RDMA Building Remote Pointer Types Building Distributed Data Structures Extending to GPUsThis Talk Background: how do we write a program for a supercomputer? Introduce PGAS Model

periods The sum of a sufficient number of independent, identically distributed random variables is approximately normally distributed. STABILIZER repeatedly randomizes layoutSTABILIZER generates a new periods The sum of a sufficient number of independent, identically distributed random variables is approximately normally distributed. STABILIZER repeatedly randomizes layoutSTABILIZER generates a new identically distributed random variables is approximately normally distributed. STABILIZER repeatedly randomizes layoutCentral Limit Theorem execution times are normally distributed The sum of a

>90000 students • Entertainment Computing Research Group • Efficiency and performance of game engines, AI, networking, VR … • Teaching: 3D Graphics, AI, Physics for games, Game Streaming, … • IFIP Hlavacs – Combining Co-Routines and Functions into a Job System - CppCon 2021 3 / 39 Creating Game Engines with C++ • Vienna Game Job System + • Graphics API Abstraction Layer + • Vienna Entity Component share main memory, can share caches • Simultaneous multithreading (x86 / x64) -> 2N virtual cores • Query number of cores: std::thread::hardware_concurrency() • AMD : Ryzen: 2-64, Epyc 4-64 • Intel: Core

Rest API and generic repositories to host any artifact. • The conan_server is a small server distributed together with the Conan client. It is a simple open-source imple- mentation, it provides the basic different configurations. • Automated/remote package generation in Travis/Appveyor server with distributed builds in CI jobs for big/slow builds. • Docker: Automatic generation of packages for several the --packages/-p or --query/-q flags. --packages allows you to explicitly declare which package gets uploaded to the server by specifying the package ID. --query accepts a query parameter, e.g. arch=armv8

Rest API and generic repositories to host any artifact. • The conan_server is a small server distributed together with the Conan client. It is a simple open-source imple- mentation, it provides the basic different configurations. • Automated/remote package generation in Travis/Appveyor server with distributed builds in CI jobs for big/slow builds. • Docker: Automatic generation of packages for several the --packages/-p or --query/-q flags. --packages allows you to explicitly declare which package gets uploaded to the server by specifying the package ID. --query accepts a query parameter, e.g. arch=armv8

Rest API and generic repositories to host any artifact. • The conan_server is a small server distributed together with the Conan client. It is a simple open-source imple- mentation, it provides the basic Documentation, Release 1.27.1 • Automated/remote package generation in Travis/Appveyor server with distributed builds in CI jobs for big/slow builds. • Docker: Automatic generation of packages for several the --packages/-p or --query/-q flags. --packages allows you to explicitly declare which package gets uploaded to the server by specifying the package ID. --query accepts a query parameter, e.g. arch=armv8

Rest API and generic repositories to host any artifact. • The conan_server is a small server distributed together with the Conan client. It is a simple open-source imple- mentation, it provides the basic different configurations. • Automated/remote package generation in Travis/Appveyor server with distributed builds in CI jobs for big/slow builds. • Docker: Automatic generation of packages for several the --packages/-p or --query/-q flags. --packages allows you to explicitly declare which package gets uploaded to the server by specifying the package ID. --query accepts a query parameter, e.g. arch=armv8

Rest API and generic repositories to host any artifact. • The conan_server is a small server distributed together with the Conan client. It is a simple open-source imple- mentation, it provides the basic Documentation, Release 1.32.1 • Automated/remote package generation in Travis/Appveyor server with distributed builds in CI jobs for big/slow builds. • Docker: Automatic generation of packages for several the --packages/-p or --query/-q flags. --packages allows you to explicitly declare which package gets uploaded to the server by specifying the package ID. --query accepts a query parameter, e.g. arch=armv8