Shulou(Shulou.com)06/03 Report--

As a qualified architect, some data must be kept in mind. When designing a system, it is very important to be able to predict the load of the system through quantitative calculation, as well as the collaborative relationship between various component units, where is the bottleneck, and how to improve it. When talking about collaboration with other teams, you will also become confident and will not make directional mistakes.

Memory cache redis concurrency: 50000 Read a data from remote distributed Redis, 0.5 ms Read 1MB data from memory, more than ten microseconds Network transmission TCPIP protocol Maximum message length (MTU): Ethernet: 1500, depending on the specific TCP implementation, the communication parties will negotiate the network transmission 2 KB data, 1 microsecond (Of course, under low load) Disk IOIDE iops of hard disk: 80~100SATA iops: 120~150SAS disk: 150~180SSD disk: 5000~10000PCIE SSD: 10000~16000 Database (mysql) concurrency: 5000 Query a record (with index) in the database, i.e.: QPS: Write a record in the database in ten milliseconds: 1 ms Executing a transaction in the database: hard to say, see the specific transaction type Database table Maximum record: 10 million, preferably not more than 20 million Web server processing requests per second Apache: 20000 times per second Nginx: 20000+ times per second Tomcat: 500~1000 times per second Open a website: a few seconds, 3~5 seconds, 5 seconds is the endurance limit

With the development speed of software and hardware in IT industry, these data are constantly updated. Therefore, this data is only for reference, and it should be calculated in combination with its own real environment.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.