Topics / Reducing synchronization

Performance topic

Reducing synchronization

Less contention between threads: lock-free structures, sharding, per-thread state, relaxed atomics and fewer barriers.

1,462proven techniques
72pitfalls: tried, then slower
618repositories
84from papers

Languages

  • C++616
  • C260
  • Rust151
  • Python108
  • CUDA97
  • Go67

Hardware

  • CPU363
  • NVIDIA GPU243
  • GPU129
  • x86125
  • ARM52

Fields

  • general539
  • GPU294
  • data & compression235
  • graphics144
  • CPU113

Where these come from

Repositories with the most techniques on this topic in the database:

torvalds/linux · 20crossbeam-rs/crossbeam · 14torvalds/linux · 13AdaptiveCpp/AdaptiveCpp · 10torvalds/linux · 10memgraph/memgraph · 10microsoft/mscclpp · 10haproxy/haproxy · 9uccl-project/uccl · 9concurrencykit/ck · 9

Ask Claude for the full answer

Each technique comes with its bottleneck, measured speedup, trade-offs, conditions and the commit or paper behind it, served to Claude through MCP.

claude mcp add --transport http adtomic https://www.adtomic.co/mcp

Connect Claude Pricing

Related topics

Pipelining and async overlapBit packing and smaller typesLazy evaluationZero-copyBranchless programming