Introduction
Little Monkey is 24 years old this year. He has just graduated from university and
is currently an intern at an Internet M company.
He spent the whole night last night, and
nothing else,
because the company newly added 500 servers last night and
asked Little Monkey and other newcomers to install the system.
Correct,
Installed 1 whole day + 1 overnight...
finally at night because of a flash of lightning
The computer room is out of power...has to
be reinstalled
Today, the little monkey found his monkey brother-data ape Wen Da
Ask him to complain
1. Cloud Computing
1.1 The essence of cloud computing: anytime, anywhere + willful deployment resources
At first, it was very inflexible to manage resources through physical machines:
- Inflexible procurement
- Inflexible granularity
- Inflexible reuse
- Inflexible operation and maintenance
In order to make it flexible and easy to use, the major server vendors have virtualized on physical machines. It is convenient to virtualize the resource environment you need through manual configuration, but the scale is also limited. Later, the cloud is finally realized on the basis of virtual machines. Computing and environments are deployed in a unified data center, and then flexibly provide customers with computing resources, network resources, and storage resources through the Internet.
1.2 Have flexibility in time and space
Cloud computing can meet your resource needs at any time. For
example, at 11 o'clock in the evening, if you want a space of 8g memory, 4 core cpu, 100g hard disk, you only need to configure it with a single click on the cloud platform, without you. Buy physical machines, install systems, and configure the network environment by yourself, and everyone's resources are virtualized and independent.
1.3 Virtualization semi-automatic, cloud computing fully automatic
Virtualization is usually to virtualize multiple systems on a physical machine, but this also requires manual operation. However, as the scale of the cluster expands, Internet companies basically start with thousands of servers, and there are tens of thousands or even more. With millions of servers, it is almost impossible for people to configure virtual machines for so many servers. In the end, machines are required to do this.
Later invented something called the scheduling algorithm, it is like a dispatch center, thousands of servers in this center which, regardless of how many virtual resource users cpu, memory, hard disk needs, it can automatically find in the center to meet the user's profile, And start the virtual computer, the user can use it directly, the whole stage is called: cloud computing
1.4 Private cloud and public cloud have their own advantages
1.5 Cloud vendor wearing an e-commerce jacket-Amazon
The original Amazon was an e-commerce website. When it was an e-commerce company, it foresees that it would encounter the spike peak scenario of Double Eleven: at a certain moment everyone rushed to buy something, and when they did it, they needed the flexibility of the cloud. Yes, tens of thousands of servers need to be deployed on Double Eleven, and these resources need to be released to do other things after Double Eleven. Therefore, Amazon develops its own cloud software based on open source virtualization technology. The cloud platform of Amazon needs to support its own e-commerce applications; while traditional cloud computing vendors are mostly IT vendors and have almost no applications of their own. Therefore, Amazon’s cloud platform is more application-friendly and has rapidly developed into the number one brand of cloud computing. A lot of money.
1.6 Iaas, Paas, Saas are optional
1. Iaas is cooking at home
Living in a residential building, you have a large kitchen. You clean the ingredients purchased from the supermarket, turn on the gas and pour rapeseed into the pot, and turn on your mobile phone to search for recipes. After 40 minutes have passed, Meimei has a meal. After eating, while cleaning the pots and dishes, I still think about "buying a dishwasher sooner or later". Gas and cooking space are already provided. This is "a perfect living facility in an industrial society."
2. Paas is to bring vegetables to a friend’s home to cook
Buy your own vegetables and go to a friend’s home to cook. The friend’s home provides kitchenware + gas + tableware. You only need to show your superb cooking skills to save time and money for each meal, and at the same time get the appreciation of your friends. This is "reuse of resources in the era of sharing economy".
3. Saas just go to the restaurant
When it's time for dinner, it's time to eat, you go to KFC downstairs. You proficiently scanned the menu with your mobile phone and quickly ordered a "hamburger with more tomato sauce". After a few minutes, the steaming burger was brought to you. After a whole gorge, you wiped your mouth and left. This is a "contemporary young man" The status quo of life".
These 3 models correspond to 3 types of cloud computing:
- The first is Infrastructure Services (Iaas) The kitchen is storage and servers, and the fuel is the network . Your company does not need to worry about servers, storage, networks, etc., because they are just like hydropower in modern urban life, and are the infrastructure of the cloud computing era.
- The second is the platform (Paas) where the cookware is the operating system and the tableware is the middleware . Your company has purchased platform services, so in addition to infrastructure, service providers will also provide you with various development and use solutions through middleware, saving time and resources.
- The third is of course software (Saas) , which is the basic level that ordinary consumers contact . The most common is all kinds of web micro-applications. You don't even need to prepare the data (food ingredients) yourself, just think about what you need, and then you can use it.
1.7 Cloud computing summary
