Topics / I/O batching

Performance topic

I/O batching

Fewer, larger I/O operations: buffered writes, io_uring, vectored reads and coalesced network packets.

604proven techniques
10pitfalls: tried, then slower
294repositories
181from papers

Languages

  • C++164
  • C119
  • Python56
  • Rust55
  • Go27
  • CUDA20

Hardware

  • CPU160
  • NVIDIA GPU59
  • GPU49
  • x8630
  • ARM14

Fields

  • data & compression236
  • general119
  • networking72
  • graphics64
  • GPU56

Where these come from

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

vgmstream/vgmstream · 8libuv/libuv · 7DynamoRIO/dynamorio · 6OSGeo/grass · 5pgbackrest/pgbackrest · 5lance-format/lance · 4postgres/postgres · 4memgraph/memgraph · 4qdrant/qdrant · 4python/cpython · 4

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

Fast math and approximationSIMD vectorizationBetter algorithmsPrecomputation and memoizationAvoiding allocations