SQL Server is a high frequently used piece of software which need to serve single requests and/or hundreds of thousands of requests in a minute. Within these different kinds of workloads Microsoft SQL Server has to handle the concurrency of tasks in a fashion manner. This demo driven session shows different scenarios where Microsoft SQL Server has to wait and manage hundreds of tasks. See, analyze and solve different wait stats due to their performance impact: – CXPACKET: when a query goes parallel – ASYNC_IO_COMPLETION: speed up IO operations (Growth / Backup / Restore) – ASYNC_NETWORK_IO: What happens if your application refuses data? – THREADPOOL starvation: crush of requests for Microsoft SQL Server – PAGELATCH_xx: How Microsoft SQL Server protects data?