Linux Lesson1 Summary

First, the basic principle of computer

1. website PC Server

The unit is in 1u=4.45cm. (2U, 4U)

2. Motherboard BIOS chip

BIOS chip is responsible for the motherboard power after the computer parts self-test, set, save, all normal after the start.

Q: Please describe the whole process of Linux booting to system login:

1, boot BIOS self-test;

2. MBR Boot

3. Grub boot Menu

4. Load Kernel Kernerl

5. Start the INIT process

6, read Inittab file, execute rc.sysinit,rc. etc script

7, start the Mingetty, enter the system login screen

3. CPU

Function: computer control and Computing Center.

Function: 1, any operation of the computer (data storage and input, program operation, screen printing, etc.) are under the control of the CPU. The CPU is more capable of determining the computational speed of a computer than any part of the computer.

2, CPU performance is determined by the number of cores and frequency of the CPU. (Frequency is the number of times a CPU can work per second)

4. HDD

Function: Hard disk is a large capacity memory

Function: When calculating the running time, the CPU and input device, the output device, the memory must exchange the data, the memory speed and the capacity is also one of the important indexes which determines the computer operation speed. For the server, the performance of the hard disk is also one of the important factors determining the performance of the website server.

5. Memory

Function: Memory is the temporary memory of the computer, it is responsible for the relay of the data, and cannot be saved permanently. Memory capacity and processing speed is also one of the important factors to determine the speed of computer operations.

Function: Memory is directly accessed by the CPU, the CPU reads the operation instruction and data from memory, and the result of processing operation is transmitted to memory.

Read-Write performance sequencing:

CPU (L1cache L2cache) > Memory > Ssd> sas> SATA

Summary:

First, the PC server brand and model

Dell Brand:

2010-2013 1u:r410 R610 2u:710

2014-2015 1u:r420/430 r620/630 2u:r720

Second, the main components of the server

1, Power: General production environment are configured with dual power supply, two-line AB Road.

A server should be configured with dual power supply, multiple hosts can realize the cluster work, do not configure dual power supply.

2, CPU computer operation and Controller

3, the memory CPU and the buffer between the disk device, when the program is running, it will be transferred to the memory execution, when the computer shutdown or the end of the process, the process is freed from memory.

Process: A program that is in progress that is invoked to execute in memory.

Program: Static files placed inside the disk, PHP Java, etc.

Daemon: Keep running processes in progress

Service: Runs software that provides functionality.

The key to optimizing websites and servers: use memory as much as possible.

The difference between Q:buffer and cache:

Buffer buffer: Buffers are designed to increase the speed of data exchange between memory and disk, and when data is transferred between memory and slow disk storage, the slower data is placed in the buffer, which accumulates to a certain amount, and is provided to memory for reading. This accelerates the efficiency of the memory;

Cache: The buffer is designed to improve the speed of data exchange between CPU and memory, the CPU reads and writes faster than the memory read/write speed, and the CPU takes a long time reading the data directly from memory, and the cache area holds the CPU just used, or partially recycled data, which improves the performance of the system.

4. Disk

Interface type: SSD SCSI IDE SATA SAS

Performance: Ssd>sas>sata

5. Raid Card

Basic function: The single disk does not fit all the data, so you need to consolidate all the disks into a disk, partition the entire disk, and then store the data.

The difference between ps:raid0, RAID1, RAID5 and RAID10:

1, RAID0 will all consolidation of a disk, the space utilization rate of 100%, side-by read and store data, so read and write performance has improved, but the redundancy capacity of 0, redundancy is fault-tolerant ability. If a disc is damaged, all discs are damaged.

2, RAID1 mirror disk, the same data on two disks, so that even if a piece of disk corruption, will not cause data loss. High cost, suitable for scenarios where data is very important. But space utilization is only 50%, resulting in half of the wasted disk space, with a redundancy capacity of 50%.

3, RAID5 using parity information to backup data. Suppose there are four disks, the four disk to find a disk in turn as backup verification information, so that even if any disk in the four disk damage, the data can be recovered. But only a piece of disk data is allowed to be lost. The total number of disks is n+3.

4, RAID10 RAID10 combines the advantages of RAID0 and RAID1; it is like the RAID0, write-side-by-read data, the capacity of the total disk capacity of the sum of the various disks, and read and write ability has a great increase; a mirrored structure inside each disk, even if any one of the disks is damaged, Data can be retrieved. The space utilization is 50%, the redundancy capacity is 100%, and at least 4+2n block disks are required.

Linux Lesson1 Summary