Profiling on My Bloghttps://thaig2pro.github.io/en/tags/profiling/Recent content in Profiling on My BlogHugo -- 0.146.0enMon, 05 Oct 2026 09:00:00 +0700I Explained a Slow Read and Wrote It Down. The Ratio That Disproved It Was Already in the Table.https://thaig2pro.github.io/en/posts/cho-cham-khong-nam-o-cho-ban-doan-cap-phat-khong-phai-giai-ma/Mon, 05 Oct 2026 09:00:00 +0700https://thaig2pro.github.io/en/posts/cho-cham-khong-nam-o-cho-ban-doan-cap-phat-khong-phai-giai-ma/A 60-version row read 8.5x slower than a 1-version row. I blamed decoding, published that, and a month later a profiler showed the cost was allocation and copying: 5213 to 708 ns.