{"type":"rich","blog_title":"KaiGai\u306e\u4ffa\u30e1\u30e2","html":"<iframe src=\"https://hatenablog-parts.com/embed?url=https%3A%2F%2Fkaigai.hatenablog.com%2Fentry%2F2019%2F11%2F01%2F170159\" title=\"Billion rows processed per second at a single-node PostgreSQL - 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>","author_url":"https://blog.hatena.ne.jp/kaigai/","description":"I have worked on benchmarking of PG-Strom at a large hardware configuration for a couple of months. Due to the server models we had, our benchmark results had been usually measured at a small 1U rack server with 1CPU, 1GPU and 3-4 NVME-SSDs, even though PG-Strom supports multi-GPUs and striping of N\u2026","title":"Billion rows processed per second at a single-node PostgreSQL","provider_url":"https://hatena.blog","provider_name":"Hatena Blog","url":"https://kaigai.hatenablog.com/entry/2019/11/01/170159","width":"100%","blog_url":"https://kaigai.hatenablog.com/","categories":[],"image_url":"https://cdn-ak.f.st-hatena.com/images/fotolife/k/kaigai/20190924/20190924132242.jpg","published":"2019-11-01 17:01:59","version":"1.0","author_name":"kaigai","height":"190"}