{"height":"190","description":"An article for none-Japanese readers....What I'm recently working on is a feature to load data blocks from NVMe-SSD to GPU using peer-to-peer DMA. It allows to bypass the CPU/RAM under a series of data loading process, thus, also allows to reduce number of expensive data copy. Once data blocks are l\u2026","title":"(EN) GpuScan + SSD-to-GPU Direct DMA","blog_title":"KaiGai\u306e\u4ffa\u30e1\u30e2","version":"1.0","categories":["GPU","SSD","PostgreSQL"],"provider_name":"Hatena Blog","type":"rich","html":"<iframe src=\"https://hatenablog-parts.com/embed?url=https%3A%2F%2Fkaigai.hatenablog.com%2Fentry%2F2016%2F09%2F08%2F003556\" title=\"(EN) GpuScan + SSD-to-GPU Direct DMA - KaiGai\u306e\u4ffa\u30e1\u30e2\" class=\"embed-card embed-blogcard\" scrolling=\"no\" frameborder=\"0\" style=\"display: block; width: 100%; height: 190px; max-width: 500px; margin: 10px 0px;\"></iframe>","blog_url":"https://kaigai.hatenablog.com/","width":"100%","provider_url":"https://hatena.blog","author_name":"kaigai","author_url":"https://blog.hatena.ne.jp/kaigai/","url":"https://kaigai.hatenablog.com/entry/2016/09/08/003556","published":"2016-09-08 00:35:56","image_url":"https://cdn-ak.f.st-hatena.com/images/fotolife/k/kaigai/20160825/20160825002243.png"}