std::simd how to express inherent parallelism efficiently via data-parallel types Dr. Matthias Kretz GSI Helmholtz Center for Heavy Ion Research CppCon ’23 @mkretz@floss.social github.com/mattkretzMotivation

Division Team name Microsoft Confidential How Visual Studio Code can help you develop more efficiently in C++ From your First Line of Code to your Largest Repo:Microsoft C++ Meet the Speakers Sinem Typescript for C++ – Herb Sutter (Keynote) • How Visual Studio Code Can Help You Develop More Efficiently in C++ – Alexandra Kemper & Sinem Akinci • Regular, Revisited – Victor Ciura Friday 6th​ •

environment, we always want to use the environment’s computing resources more cost-effectively and efficiently. Cluster managers such as K8S and Yarn manage the cluster compute resources, divided into different conversely, when the engine is doing chubby tasks, we should be able to get and use more resources more efficiently. On the one hand, we need to rely on the resource manager’s capabilities for efficient resource 6.2. Deploying Kyuubi 63 Kyuubi, Release 1.3.0 could be a bad idea for using the resources efficiently. In this case, we would prefer that such an enormous task be done more slowly in a limited amount

environment, we always want to use the environment’s computing resources more cost-effectively and efficiently. Cluster managers such as K8S and Yarn manage the cluster compute resources, divided into different conversely, when the engine is doing chubby tasks, we should be able to get and use more resources more efficiently. On the one hand, we need to rely on the resource manager’s capabilities for efficient resource 6.2. Deploying Kyuubi 63 Kyuubi, Release 1.3.0 could be a bad idea for using the resources efficiently. In this case, we would prefer that such an enormous task be done more slowly in a limited amount

objects of a specific data type ● ^So what exactly does this mean, and how could we do this efficiently? ● If an object is a page in a book ● then a ‘pointer’ would be the index in the back of the objects of a specific data type ● ^So what exactly does this mean, and how could we do this efficiently? ● If an object is a page in a book ● then a ‘pointer’ would be the index in the back of the objects of a specific data type ● ^So what exactly does this mean, and how could we do this efficiently? ● If an object is a page in a book ● then a ‘pointer’ would be the index in the back of the

portability. These features of modern programming systems are among the most difficult to implement efficiently, even individually. — 1984 Smalltalk upward funargs (read)Efficient implementation of the portability. These features of modern programming systems are among the most difficult to implement efficiently, even individually. — 1984 Smalltalk (read)Smalltalk Efficient implementation of the Smalltalk-80 portability. These features of modern programming systems are among the most difficult to implement efficiently, even individually. — 1984 It’s interesting to see what authors call out about their work.

environment, we always want to use the environment’s computing resources more cost-effectively and efficiently. Cluster managers such as K8S and Yarn manage the cluster compute resources, divided into different conversely, when the engine is doing chubby tasks, we should be able to get and use more resources more efficiently. On the one hand, we need to rely on the resource manager’s capabilities for efficient resource queue/namespace and occupy them for a considerable time, which could be a bad idea for using the resources efficiently. In this case, we would prefer that such an enormous task be done more slowly in a limited amount

environment, we always want to use the environment’s computing resources more cost-effectively and efficiently. Cluster managers such as K8S and Yarn manage the cluster compute resources, divided into different conversely, when the engine is doing chubby tasks, we should be able to get and use more resources more efficiently. On the one hand, we need to rely on the resource manager’s capabilities for efficient resource queue/namespace and occupy them for a considerable time, which could be a bad idea for using the resources efficiently. In this case, we would prefer that such an enormous task be done more slowly in a limited amount

environment, we always want to use the environment’s computing resources more cost-effectively and efficiently. Cluster managers such as K8S and Yarn manage the cluster compute resources, divided into different conversely, when the engine is doing chubby tasks, we should be able to get and use more resources more efficiently. On the one hand, we need to rely on the resource manager’s capabilities for efficient resource queue/namespace and occupy them for a considerable time, which could be a bad idea for using the resources efficiently. In this case, we would prefer that such an enormous task be done more slowly in a limited amount

environment, we always want to use the environment’s computing resources more cost-effectively and efficiently. Cluster managers such as K8S and Yarn manage the cluster compute resources, divided into different conversely, when the engine is doing chubby tasks, we should be able to get and use more resources more efficiently. On the one hand, we need to rely on the resource manager’s capabilities for efficient resource queue/namespace and occupy them for a considerable time, which could be a bad idea for using the resources efficiently. In this case, we would prefer that such an enormous task be done more slowly in a limited amount