1. What challenges does traditional IT operation and maintenance face?
1. Slow business launch: new businesses are slow to launch due to reasons such as long procurement and deployment cycles
2. Difficulty in expansion: the centralized architecture has poor scalability, and vertical expansion can only increase the processing performance of a single machine. 3. Poor reliability: traditional hardware devices exist in isolation, and reliability can only be realized on the software side
4. Complex life cycle management: There are many types of equipment in the enterprise and many manufacturers, resulting in very complicated management and high skill level requirements for operation and maintenance personnel
5. Delay caused by I/O bottleneck: the performance of a single device is limited
6. TOC remains high: the overall utilization rate of the equipment is not high, and the total cost support of the enterprise remains high
Second, the characteristics of cloud computing
1. On-demand self-service: Consumers can deploy cloud services according to their needs without too much manual interaction with suppliers. 2. Extensive network access: There are not too many restrictions on the network environment, and various clients can be used for access. 3. Resource pooling: Suppliers centralize the supply of computing resources to form a resource pool, and users can arbitrarily expand the corresponding services in the resource pool. 4. Rapid deployment and flexible expansion: cloud computing can quickly and flexibly expand capacity, and for users, it is to realize the use of resources by paying
5. Measurable services: Cloud service charges are based on the resources actually used by users
3. Virtualization
1. The concept in the virtual machine:
Guest OS: virtual machine operating system
Guest Machine: virtualized virtual machine
Hypervisor: virtualization layer (VMM)
Host OS: the operating system running on the physical machine
Host Machine: physical machine
2. Virtualization is classified according to architecture:
(1) Residential virtualization: Refers to the virtualization management software as an ordinary application program on the system, and then creates a virtual machine through it to share the underlying service resources. Such as Vmware workstation virtualbox
Advantages: Simple and easy to implement Disadvantages: Installing and running the program depends on the hardware support of the host operating system, which has high management overhead and high performance consumption
(2) Bare metal virtualization: refers to the virtualization layer running directly on the physical hardware. The virtualization layer of the hostless operating system mainly realizes two basic functions: 1. Identify, capture and respond to CPU instructions issued by the virtual machine 2. Responsible for Process virtual machine queues and scheduling, and return the processing results of physical hardware to the corresponding virtual machines
Advantages: The virtual machine does not depend on the operating system, and supports multiple operating systems and applications. Disadvantages: It is difficult to develop the virtual layer kernel Vmware vsphere Vmware ESXI Vmware Vcenter huawei fusioncompute CAN VRM
(3) Operating system virtualization: It means that there is no independent virtualization layer, and the host operating system itself is responsible for directly allocating hardware resources among multiple virtual servers and making these servers independent of each other.
Advantages: simple, easy to implement, lower management overhead Disadvantages: poor isolation, virtual machines must use the same operating system docker
(4) Hybrid virtualization: use the host's operating system like host virtualization, but instead of placing the hypervisor on the host's operating system, several kernel-level drivers are inserted into the kernel of the host's operating system. This driver As a manager for virtual hardware
Advantages: High performance, supports multiple operating systems Disadvantages: The underlying hardware needs to support virtualization extensions
3. The characteristics of the container:
Packaging: Packaging software into standardized units for development, migration, and deployment
Isolation: computing, network, storage and other resources are isolated from each other
Efficiency: lightweight, fast start and stop, fast deployment and migration
Clear division of responsibilities: developers can focus on code, and operation and maintenance personnel can focus on basic environment configuration
4. Network
1. The network is a bridge for communication between devices and virtual machines, so the network is indispensable in the ICT infrastructure.
Unicast communication: is a one-to-one communication model
Multicast communication: It is a one-to-many communication model. The information of multicast communication is directed to a multicast group. A multicast group is a group with the same characteristics.
Broadcast communication: is a one-to-all communication model
2. Routing (routing table): the process of guiding data forwarding between different network devices and finally reaching the destination. A device capable of routing is usually called a router
Default gateway: The gateway is the exit of the subnet. Any data that needs to be forwarded to other subnets needs to be forwarded to the gateway for processing. Therefore, gateway devices generally connect to multiple networks and have routing functions. Common gateway devices (routers, firewalls) , Layer 3 switch).
VLAN: virtual local area network, each vlan is a logical group, users can be isolated inside the LAN through vlan technology, users of the same vlan can directly communicate on the second layer, and devices of different vlans are logically isolated. VLAN is not restricted by physical location, and can realize isolation between different businesses
Router: Used to connect different networks and direct data packets to forward data between different networks. The router will process and forward the data according to the routing table. The routing table can be manually configured by the administrator (static routing), or calculated by a dynamic protocol (common dynamic routing protocols: ospf, rip, isis, bgp) Router with few ports
Switch: Working at the data link layer in the OSI reference model, it is a network device used for electrical signal forwarding, allowing any two network nodes to have their own corresponding circuit paths. The switch has high-density ports, which can allow a large number of terminal devices to be connected to the network. In the traditional network, Layer 2 switches mainly use VLANs to isolate the network layer.
OSI reference model: application layer session layer presentation layer transport layer network layer data link layer physical layer
TCP/IP peer-to-peer model: application layer transport layer network layer data link layer physical layer
Layer 3 switch: A layer 3 switch is a switch with IP addressing function (routing function), which can replace routers in some scenarios to provide enterprises with data forwarding between different networks
Network card (NIC): The main function of the network card is to connect the media of different devices, because with the network card, it is possible to communicate between devices. In the network, the reliability of the network card and the performance of the network can be improved by binding the network card.
OVS (open VSwitch): is an open source virtual switch based on software
EVS (Enhance Vswitch): Enhanced virtual switch, based on OVS forwarding technology, improves IO performance
DVS (distributed vswitch): Like a physical switch, it can provide a network path between virtual machines, can construct a topology with a real switch in a physical network, and can build a richer and more flexible virtual network.
VPC (Virtual Private Cloud): Allows users' virtual machines to be isolated through VPC.
NAT gateway: a gateway device that provides public network NAT and private network NAT
Elastic public IP: EIP can provide elastic public IP address for ECS
5. Storage
The role of the storage medium: data storage is a temporary file generated during the processing of the data stream or information that needs to be searched during the processing, and the data is recorded in a certain format on the internal or external storage medium of the computer.
1. Block storage: Block storage means that the entire mapping of raw disk space to the server is highly versatile and widely used. It can be used for data storage in most common business scenarios. Block storage cannot be used directly in the operating system. The delay in using block storage after formatting the block storage and creating the file system is the lowest, which is suitable for high IO usage environments
2. File storage: Similar to a shared folder, a file system already exists, and users can access the folder to realize file storage. Usually, protocols such as NFS and CIFS can be applied.
NFS: Network file system, generally used for sharing files between Linux systems
CIFS: Common network file system, the general client is a Windows device, it will be more secure when used
3. Object storage: It adopts a brand-new storage architecture, so that it has both the characteristics of high-speed direct disk access of block storage and the distributed sharing characteristics of file storage. Therefore, it can store user data directly like file storage , but its performance is better than file storage. It can directly store user data like file storage, but its performance is better than file storage. Object storage is suitable for big data, IOT, backup and archiving.