boshjerns
- 16 skills
- 0 followers
- 6 hours ago last updated
- ▌ Performance Cpp · boshjerns bundleUse Jeff Dean and Sanjay Ghemawat's canonical C++-specific performance guidance. Use when selecting Abseil or Google C++ containers, limiting Status or StatusOr overhead, or reviewing concrete C++ type choices.
- ▌ Performance Hints · boshjerns bundleRoute performance reviews, optimization work, profiling investigations, API or data-structure design, allocation reduction, concurrency tuning, Protocol Buffer work, and C++ performance questions to the relevant section of Jeff Dean and Sanjay Ghemawat's canonical Abseil Performance Hints article. Use for broad or multi-area performance work when more than one focused performance subskill may apply.
- ▌ Performance Protobuf · boshjerns bundleUse Jeff Dean and Sanjay Ghemawat's canonical Protocol Buffer performance guidance. Use when reviewing protobuf schema choices, parsing, serialization, ownership, field access, or message allocation and lifetime.
- ▌ Performance Thinking · boshjerns bundleApply the general performance philosophy in Jeff Dean and Sanjay Ghemawat's canonical Performance Hints. Use when deciding whether performance should influence a design, identifying hot paths, reviewing library defaults, or balancing speed against readability and complexity.
- ▌ Performance Code Size · boshjerns bundleUse Jeff Dean and Sanjay Ghemawat's canonical code-size guidance. Use when instruction-cache pressure, excessive inlining, template expansion, or repeated container operations may increase binary size or degrade performance.
- ▌ Performance Algorithms · boshjerns bundleUse Jeff Dean and Sanjay Ghemawat's canonical algorithmic-improvement guidance. Use when asymptotic complexity, redundant searches, data-structure choice, or a fundamentally better algorithm may dominate lower-level optimizations.
- ▌ Performance API Design · boshjerns bundleUse Jeff Dean and Sanjay Ghemawat's canonical API performance guidance. Use when designing or reviewing bulk APIs, view types, caller-provided storage, precomputed arguments, or thread-compatible versus thread-safe interfaces.
- ▌ Performance Estimation · boshjerns bundleUse Jeff Dean and Sanjay Ghemawat's canonical estimation guidance for back-of-the-envelope performance calculations. Use when comparing designs before implementation, estimating latency or resource cost, reasoning about operation counts, or deciding what is likely to dominate.
- ▌ Performance Allocations · boshjerns bundleUse Jeff Dean and Sanjay Ghemawat's canonical allocation guidance. Use when reducing heap traffic, copies, container growth, temporary construction, or repeated allocation in hot paths.
- ▌ Performance Measurement · boshjerns bundleUse Jeff Dean and Sanjay Ghemawat's canonical measurement guidance for benchmarks, profiling, flat profiles, and optimization validation. Use when collecting evidence, selecting profiling tools, interpreting profiles, or proving that a change improved performance.
- ▌ Performance Parallelism · boshjerns bundleUse Jeff Dean and Sanjay Ghemawat's canonical parallelization and synchronization guidance. Use when tuning threads, locks, contention, critical sections, sharding, SIMD, false sharing, context switches, pipelines, channels, or lock-free designs.
- ▌ Performance Case Studies · boshjerns bundleRead Jeff Dean and Sanjay Ghemawat's canonical examples of change lists that combine multiple performance techniques. Use when looking for worked optimization patterns, evaluating a multi-part refactor, or comparing a proposed change with real compound examples.
- ▌ Performance Bulk Operations · boshjerns bundleUse Jeff Dean and Sanjay Ghemawat's canonical bulk-operations guidance. Use when multiple transformations can be fused or batched to reduce passes, round trips, dispatch overhead, or intermediate materialization.
- ▌ Performance Further Reading · boshjerns bundleRead the performance books, articles, talks, and other resources recommended by Jeff Dean and Sanjay Ghemawat in their canonical Performance Hints article. Use when the user wants deeper study, primary references, or resources beyond the individual hints.
- ▌ Performance Unnecessary Work · boshjerns bundleUse Jeff Dean and Sanjay Ghemawat's canonical guidance on avoiding unnecessary work. Use when adding fast paths, hoisting loop invariants, deferring computation, specializing code, caching, helping the compiler, or reducing statistics and logging overhead.
- ▌ Performance Memory Representation · boshjerns bundleUse Jeff Dean and Sanjay Ghemawat's canonical memory-representation guidance. Use when tuning cache locality, object size, pointer density, container layout, batched or inline storage, maps, sets, bit vectors, or arenas.