 Servers are computers that provide services to users, and are generally divided into file servers, database servers, and application program servers.

1. The main features of the server include:

▫ R: Reliability—reliability, how long it can continue to operate normally;

▫ A: Availability—availability, the percentage of system uptime and usage time;

▫ S: Scalability—scalability, including two aspects, one is the scalability of the hardware, and the other is the ability of the software to support the operating system;

▫ U: Usability—ease of use, the hardware and software of the server are easy to maintain and repair;

▫ M: Manageability—manageability, which can monitor the running status of the server, alarm, and automatically and intelligently deal with some faults.

2. Server application deployment architecture:

C/S: Abbreviation for Client/Server. Also commonly referred to as client/server architecture.

The client handles the user's front-end interface and interactive operations, and the server handles the background business logic and request data, which greatly improves the communication speed and communication efficiency at both ends.

B/S: Abbreviation for Browser/Server. Also commonly referred to as browser/server architecture.

In the B/S architecture, users only need to install a browser, and the application logic is concentrated on the server and middleware, which can improve data processing performance.

3. Server classification method - hardware form

Mainframe Minicomputer Tower Server Blade Server Rack Server

4. Server Classification - Service Scale

Entry-level server equivalent to "PC server"

Workgroup server Low-end server, providing small-scale (about 50 clients) services

Department-level server Mid-range server, providing services for about 100 clients

Enterprise-class server High-end server, used for more than hundreds of client access

5. Server hardware structure

Take the hardware structure of Huawei TaiShan 200 server as an example

Chassis 2 Motherboard 3 Memory 4 CPU 5 CPU Heat Sink 6 Power Supply 7 Fan 8 Hard Disk 9 Air Duct Cover

5.1 CPU Definition and Composition

Definition of CPU

 CPU: Central Processing Unit, central processing unit, is the computing core and control core of a computer.

 CPU, internal memory and input/output devices are the three core components of an electronic computer.

 The main function of the CPU is to interpret computer instructions and process data in computer software.

The composition of the CPU

 CPU consists of logic operation unit, control unit and storage unit.

The CPU is a server with strict requirements such as fast throughput of large amounts of data, super stability, and long-term operation. The "brain" of the computer is the primary indicator of server performance

Controller: The computer controls the whole machine according to the program stored in advance, and the program refers to the instruction sequence that can realize a certain function. The controller is the mechanism that issues commands to various logic circuits according to the instructions. It is the command of the computer. The center controls the work of the entire CPU and determines the automation of the computer's running process.

Arithmetic unit: A component in a computer that performs various arithmetic and logical operations. The basic operations of the arithmetic unit include four arithmetic operations of addition, subtraction, multiplication, and division, logic operations such as AND, OR, NOT, and XOR, as well as operations such as shift, comparison, and transfer, and are also called arithmetic logic components.

Register: The main function of the register is to temporarily store the data and operation results involved in the operation, and has the functions of receiving data, storing data and outputting data.

CPU frequency

main frequency

 The main frequency is also called the clock frequency, and the unit is megahertz (MHz) or gigahertz (GHz), which is used to indicate the speed of CPU operation and data processing.

FSB

 FSB is the base frequency of the CPU, and the unit is MHz. The FSB of the CPU determines the operating speed of the entire motherboard.

bus frequency

 Bus frequency directly affects the speed of data exchange between CPU and memory.

multiplication factor

 Multiplication factor refers to the relative proportional relationship between CPU main frequency and FSB.

5.2 Memory

memory definition

 Storage can be divided into main storage and auxiliary storage according to usage. Main memory, also known as internal memory (memory for short), is a storage space that the CPU can directly address.

 The role of the memory is to temporarily store the calculation data in the CPU and the data exchanged with the external memory such as the hard disk.

 Memory is one of the important components in the computer, it is a bridge to communicate with the CPU.

 The memory is composed of memory chips, circuit boards, gold fingers and other parts.

The memory is a component used to store programs and data. For a computer, only with memory can it have a memory function and ensure normal operation.

All programs in the computer run in the memory, so the performance of the memory has a great impact on the computer. The memory is composed of memory chips, circuit boards, gold fingers and other parts.

Server memory slots and configuration principles:

▫ The same server must use the same type of DIMM.

▫ At least one memory module must be configured on the memory slot corresponding to the CPU.

▫ The best memory performance can be achieved when the server is configured with fully balanced memory sticks. Unbalanced configurations can degrade memory performance and are therefore not recommended.

5.3 Hard Disk Introduction

The hard disk is the main storage device of the computer.

The hard disk interface is the connecting part between the hard disk and the host system, and its function is to transfer data between the hard disk cache and the host memory. Different hard disk interfaces determine the connection speed between the hard disk and the computer, and directly affect the running speed of programs and system performance.

Mainstream Speed (RPM) Serial/Parallel Mainstream Capacity (TB) MTBF (h)

SATA 7,200 Serial 1T/2T/3T 1,200,000

SAS 15,000/10,000 Serial 1T/2T/3T 1,600,000

NL-SAS 7,200 Serial 2 T/3 T/4 T 1,200,000

SSD NA Serial 0.6T/0.8T/1.2T/1.6T 2,000,000

SATA3.0 achieves the highest data transfer rate of 600 MB/s. The annual failure rate of SATA hard drives is about 2%.

SAS is designed for high-performance enterprise needs and is compatible with SATA hard drives. It can provide a transmission rate of 3.0 Gb/s to 6.0 Gb/s, and plans to reach 12.0 Gb/s in the future. The annual failure rate of SAS hard drives is less than 2%.

MTBF: Mean Time Between Failure, the average working time between failures. • In terms of price, generally speaking, SATA and NL-SAS hard drives are cheaper, SAS hard drives are more expensive, and SSDs are the most expensive.

5.4 RAID card

RAID card is also called disk array card, or array card for short.

The role of the RAID card:

 Several hard disk drives can be combined into a whole system managed by the array controller according to certain requirements.

 Can improve the performance and reliability of the disk subsystem.

RAID definition

 RAID: Redundant Array of Independent Disks, Redundant Array of Independent Disks, formerly known as Redundant Array of Inexpensive Disks (Redundant Array of Inexpensive Disks), referred to as Disk Array. Using virtualization storage technology to combine multiple hard disks into one or more hard disk array groups, the purpose is to improve performance or data redundancy, or both at the same time.

RAID Hot Standby and Reconfiguration Concepts

Definition of Hot Spare

 When a hard disk in the redundant RAID group fails, without interfering with the normal use of the current RAID system, another normal spare hard disk in the RAID system will automatically replace the failed hard disk to ensure the redundancy of the RAID system in time.

Hot standby is generally divided into two types

 Global: The spare hard disk is shared by all redundant RAID groups in the system.

 Dedicated: The spare hard disk is dedicated to a redundant RAID group in the system.

Data verification, which uses redundant data to detect and repair data errors. Redundant data is usually calculated using algorithms such as Hamming codes and XOR operations. Using the parity function can greatly improve the reliability, high performance and fault tolerance of the disk array. However, data verification needs to read data from multiple places and perform calculations and comparisons, which will affect system performance.

Generally speaking, RAID cannot be used as an alternative to data backup, and it cannot do anything about data loss caused by non-disk failures, such as viruses, man-made destruction, accidental deletion, etc. The data loss at this time is relative to the operating system, file system, volume manager or application system. For RAID, the data is intact and no loss occurs. Therefore, data protection measures such as data backup and disaster recovery are very necessary, and complement each other with RAID to protect the security of data at different levels and prevent data loss.

Realization of RAID - hardware method

Hardware-based RAID is implemented using a hardware RAID adapter card.

Hardware RAID can be divided into built-in card type and external independent disk array.

The RAID card integrates a processor, which can control the RAID storage subsystem independently of the host. Because it has its own independent processor and memory, the RAID card can calculate the parity information and complete file positioning by itself, reducing the CPU computing time of the host computer and improving the parallel data transmission speed.

Realization of RAID - software method

definition

 Software RAID refers to the realization of corresponding RAID functions by installing software in the operating system.

features

 Software RAID does not require an expensive RAID controller card and provides an inexpensive solution.

 The RAID function is completely performed by the CPU, and the CPU of the host is heavily occupied, such as a large number of exclusive OR (XOR) operations in RAID 5.

The following functions cannot be provided in software RAID:

▫ Hard disk hot swap;

▫ Hard disk hot backup;

▫ Remote array management;

▫ Bootable array support;

▫ Realize the array configuration on the hard disk;

▫ SMART hard drive support.

Implementation of RAID - Method Comparison

5.5 Network card definition and function

Network card definition

 Network card, also known as network adapter or network interface card NIC (Network Interface Card), is the most basic, most important and indispensable connection device in the computer network system, and the computer is mainly connected to the network through the network card.

Main functions of network card

 represents a fixed network address

 Data sending and receiving

 Encapsulation and decapsulation of data

 Link management

 Encoding and decoding

Huawei server NIC

Onboard network card

 It is integrated in the PCH chip of the server motherboard and cannot be replaced.

 Provides 2 GE electrical ports + 2 10G optical/electrical ports, and the integrated network card does not occupy the PCIe slot.

PCIe standard card

 Huawei has self-developed and outsourced PCIe standard network cards, which can be configured in standard PCIe slots.

Flexible IO card

 Huawei self-developed network card, non-standard PCIe card form, can only be used for Huawei rack servers.

Mezz card

 The dedicated Mezz card for blade server computing nodes can only be used for Huawei E9000 blade servers.

PCIe (PCI-Express) is the third-generation I/O bus after the ISA and PCI bus, namely 3GIO.

The main advantages are high data transmission rate, strong anti-interference ability, long transmission distance and low power consumption.

PCIe standard card: For Huawei servers, it refers to the network card in the PCIe slot.

5.6 Power supply and fans

Support the electrical load of the server

Supports redundancy to protect against power failure

 Fault warning and prevention

 Preventive maintenance before failure

 Keep the server up and running

The power subsystem includes:

 Intelligent power supply

 Fan

Redundant power supplies and fans

Power redundancy features:

▫ 1+1, at this time each module bears 50% of the output power, when one module is pulled out, the other module bears 100% of the output power.

▫ 2+1, there are three modules, each module is responsible for 1/3 of the output power, one module is pulled out, and the remaining two modules each bear 50% of the output power.

6. Server key technology

6.1 What is IPMI

IPMI definition

IPMI: Intelligent Platform Management Interface, intelligent platform management interface, is an open standard hardware management interface specification, which defines a specific method for embedded management subsystems to communicate.

 IPMI information is communicated through the baseboard management controller (BMC), which is managed using low-level hardware intelligence instead of the operating system.

6.2 Introduction to BMC

BMC definition

 BMC: Baseboard Management Controller, the main board management control unit, is the core of the IPMI specification, responsible for signal collection, processing, storage of various sensors, and monitoring of the operating status of various devices. The BMC provides information such as the hardware status and alarms of the managed objects to the shelf management board, thereby realizing the device management function of the managed objects by the shelf management module.

BMC mainly implements the following functions:

▫ Remote control;

▫ Alarm management;

▫ Status detection;

▫ Equipment information management;

▫ Heat dissipation control;

▫ Support ipmitool tool;

▫ Support web interface management;

▫ Support centralized account management.

6.3 Introduction to iBMC

Huawei Intelligent Baseboard Management Controller (iBMC) is a server embedded management system with completely independent intellectual property rights and oriented to the whole life cycle of servers.

Introduction to BIOS

BIOS: Basic Input/Output System, basic input and output system.

BIOS is actually a set of programs solidified into the computer, providing the computer system with the lowest and most direct hardware control service, and solving the real-time needs of the hardware.

Functions of BIOS:

 Detect and initialize hardware

 Operating system boot

 Advanced Power Management

The BIOS is the bridge between the system kernel and the hardware layer.

BIOS features:

▫ Software upgrade, loading and loading functions;

▫ Basic OAM functions;

▫ Serial port management function;

▫ Fault recovery function;

▫ ECC management function;

▫ Hardware diagnostic function.

Which of the following descriptions about Huawei server NICs are correct?

A. The onboard network card is integrated in the PCH chip of the server motherboard and cannot be replaced

B. The PCIe standard network card developed by Huawei can be configured in a standard PCIe slot

C. The flexible IO card is integrated on the server panel and can be used for front-end business connections

D. Mezz cards can be used in Huawei rack servers

▫ AB

2. The main board management control unit (BMC) is the core of the IPMI specification, responsible for signal collection, processing, storage of various sensors, and monitoring of the operating status of various devices.

A. correct

b. error

Reference answer:

▫ A

Huawei Cloud Computing HCIA (1) - Server

cloud computing

server definition