Workshop: Advanced SYCL Concepts for Heterogenous Computing
The workshop covered advanced concepts and features of the latest SYCL
specifications, including simplified code implementation for
heterogeneous computing, an overview of memory models involving
buffers/accessors and unified shared memory (USM), and an understanding
of simplified reduction with live hands-on sample codes on Intel®
DevCloud.
Here’s a detailed outline of the topics that will be covered in the
workshop:
- Learn pointer-based memory management for heterogeneous computing
using unified shared memory.
- Understand the implicit and explicit ways of moving memory using
unified shared memory and handling data dependency between kernel
executions.
- Understand the advantages of using Subgroups in SYCL programming
- Take advantage of Subgroup group algorithms and understand how
Subgroup Shuffle operations can avoid explicit memory operations.
- Use SYL reduction to simplify reduction with parallel kernels.
- Take advantages, reduce function to reduce sub_group and work-group level.