<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>System Design on Boyang Yue</title><link>http://blog.boyangyue.com/tags/system-design/</link><description>Recent content in System Design on Boyang Yue</description><generator>Hugo</generator><language>en-us</language><copyright>A Good Year Ahead</copyright><lastBuildDate>Sun, 11 Sep 2022 11:00:00 +0900</lastBuildDate><atom:link href="http://blog.boyangyue.com/tags/system-design/index.xml" rel="self" type="application/rss+xml"/><item><title>Two Measures of Fast: Throughput and Latency</title><link>http://blog.boyangyue.com/2022/09/two-measures-of-fast-throughput-and-latency/</link><pubDate>Sun, 11 Sep 2022 11:00:00 +0900</pubDate><guid>http://blog.boyangyue.com/2022/09/two-measures-of-fast-throughput-and-latency/</guid><description>&lt;p&gt;A service can double the requests it completes per second while every individual request gets slower. There is no contradiction, because the two numbers answer different questions: &lt;strong&gt;latency&lt;/strong&gt; is the time one operation takes, from request to response, while &lt;strong&gt;throughput&lt;/strong&gt; is the rate at which the system as a whole completes operations. Much of performance engineering consists of keeping the questions apart: knowing which number a change moves, and at what cost to the other.&lt;/p&gt;</description></item></channel></rss>