Three dimensional version of dream journey to the West
Publish: 2021-05-17 03:36:17
1. To understand the requirements of cloud computing for servers, we must first have a certain understanding of cloud computing. Now there is no clear definition of cloud computing server, and there is no unified standard. Compared with traditional servers, traditional servers include processor moles, storage moles, network moles, power supplies, fans and other devices, except for instrial servers, such as Yanxiang instrial servers, which are between cloud servers and traditional servers
cloud server focuses on high-performance throughput computing power, and focuses on the sum of work over a period of time. Therefore, the architecture of cloud server is very different from that of traditional server
in architecture, cloud server architecture includes cloud processor mole, network processing mole, storage processing mole and system component mole. The advantage of this architecture makes the cloud server can greatly improve the utilization, using multiple cloud processors to complete the system design, introcing the concept of low-power management to complete the centralized rendancy management of the system, and saving a lot of repeated hardware in the system
the cloud server generally includes the sum of all the calculation of thread, core, processor, network, accelerator and other functional units. Therefore, cloud computing generally has a huge amount of data input or massive working set. So what capabilities should the server have? Let's talk about it in detail<
high density
high density and low cost are basically the basic requirements of cloud server. According to the characteristics of cloud computing center, cloud server should be consistent with the characteristics of high density, low power consumption and low cost of cloud computing center, that is, it is mainly oriented to large-scale deployment of cloud applications. High density servers can rece latency and improve response speed. At present, high-density servers are mainly divided into multi rack and blade servers
virtualization
the ability of server virtualization can directly affect the effect of cloud computing. The server virtualization technology realizes the real-time migration of some virtual machines from high load nodes to low load nodes, merges multiple low load virtual machines into one physical node, and closes the rendant idle physical nodes, so as to improve the utilization efficiency of resources, balance the load, and guarantee the performance of upper layer applications, At the same time, it also achieves the purpose of recing energy consumption
therefore, cloud computing can realize the deployment and configuration of virtual machines by using server virtualization technology. Through the rapid deployment and real-time migration of virtual machines, the efficiency of the system can be greatly improved, and the scalability of the software system can be realized through the adjustment of virtual machine resources, so as to ensure that the system can quickly recover from failure and continue to provide services, The reliability and stability of the system are improved. Therefore, when purchasing a cloud server, the support of server hardware virtualization is an important factor to consider the server<
scale out
according to the first feature of cloud computing, "super large scale", the scale out capability of cloud computing server becomes crucial. It can provide higher database and better scalability for such a large number of file access, and provide a more efficient, more secure and more simplified way for the whole cloud computing center, It ensures the flexibility of cloud data center
parallel computing
cloud computing, in a certain form, is the technical route of distributed computing, parallel computing, grid computing, etc. conceptually, cloud computing can be regarded as the organic combination of "Storage Cloud" and "computing cloud", and Computing cloud refers to parallel computing. Therefore, cloud computing can be regarded as the combination of "Storage Cloud" and "computing cloud", The infrastructure of cloud computing is to ensure parallel computing.
cloud server focuses on high-performance throughput computing power, and focuses on the sum of work over a period of time. Therefore, the architecture of cloud server is very different from that of traditional server
in architecture, cloud server architecture includes cloud processor mole, network processing mole, storage processing mole and system component mole. The advantage of this architecture makes the cloud server can greatly improve the utilization, using multiple cloud processors to complete the system design, introcing the concept of low-power management to complete the centralized rendancy management of the system, and saving a lot of repeated hardware in the system
the cloud server generally includes the sum of all the calculation of thread, core, processor, network, accelerator and other functional units. Therefore, cloud computing generally has a huge amount of data input or massive working set. So what capabilities should the server have? Let's talk about it in detail<
high density
high density and low cost are basically the basic requirements of cloud server. According to the characteristics of cloud computing center, cloud server should be consistent with the characteristics of high density, low power consumption and low cost of cloud computing center, that is, it is mainly oriented to large-scale deployment of cloud applications. High density servers can rece latency and improve response speed. At present, high-density servers are mainly divided into multi rack and blade servers
virtualization
the ability of server virtualization can directly affect the effect of cloud computing. The server virtualization technology realizes the real-time migration of some virtual machines from high load nodes to low load nodes, merges multiple low load virtual machines into one physical node, and closes the rendant idle physical nodes, so as to improve the utilization efficiency of resources, balance the load, and guarantee the performance of upper layer applications, At the same time, it also achieves the purpose of recing energy consumption
therefore, cloud computing can realize the deployment and configuration of virtual machines by using server virtualization technology. Through the rapid deployment and real-time migration of virtual machines, the efficiency of the system can be greatly improved, and the scalability of the software system can be realized through the adjustment of virtual machine resources, so as to ensure that the system can quickly recover from failure and continue to provide services, The reliability and stability of the system are improved. Therefore, when purchasing a cloud server, the support of server hardware virtualization is an important factor to consider the server<
scale out
according to the first feature of cloud computing, "super large scale", the scale out capability of cloud computing server becomes crucial. It can provide higher database and better scalability for such a large number of file access, and provide a more efficient, more secure and more simplified way for the whole cloud computing center, It ensures the flexibility of cloud data center
parallel computing
cloud computing, in a certain form, is the technical route of distributed computing, parallel computing, grid computing, etc. conceptually, cloud computing can be regarded as the organic combination of "Storage Cloud" and "computing cloud", and Computing cloud refers to parallel computing. Therefore, cloud computing can be regarded as the combination of "Storage Cloud" and "computing cloud", The infrastructure of cloud computing is to ensure parallel computing.
2. Depending on the platform,
I platform, you can consider the Xeon e31230
motherboard. You can use a board with al u, 8 single memory 4G,
a platform, you can use a Haolong 2000 or so, a new 12 core, 5000 or so, just like I use the board,
other things are the same as I use,
mainly depends on how much your server processes,
to judge
I platform, you can consider the Xeon e31230
motherboard. You can use a board with al u, 8 single memory 4G,
a platform, you can use a Haolong 2000 or so, a new 12 core, 5000 or so, just like I use the board,
other things are the same as I use,
mainly depends on how much your server processes,
to judge
3.
bus line: Rail Transit Line 2 → rail transit line 3, the whole journey is about 5.6km
1. Take Rail Transit Line 2 from Hankou railway station, pass 1 station, reach Fanhu station
2, walk about 280m, transfer to rail transit line 3
3, take rail transit line 3, pass 3 stations, reach Huiji 2nd Road Station
4, walk about 630m, Arrive at Wuhan Conference Center
4. If the cloud desktop with traditional VDI architecture is used, the configuration of back-end server should be at least 2.3ghz 2c4t * 40 = 80c160t, and if the memory is 4G * 40 = 160g, 160g memory is required. This kind of configuration costs a lot to the server. In particular, VDI, as a traditional cloud desktop architecture, has high requirements for the back-end server hardware performance, as well as high dependence on the network bandwidth environment, which brings high it costs to customers, and the user experience is also poor. Considering that the performance configuration of the existing 40 thin client terminals on your side is still very high, why not use the cloud desktop architecture of front-end computing mode to make full use of the performance of the 40 thin clients, so as to save the investment cost of back-end servers. At present, there is a kind of cloud desktop technology architecture called voi. The emergence of voi solves the problems that VDI can't solve to a certain extent. For example, taking the electronic classroom in the current university campus as an example, it costs at least 100000 yuan to carry a server with 100 terminal classrooms. However, in the same scenario, the cost of the server using voi solution is less than one fifth of its cost. In the electronic classroom and some institutions engaged in graphics and image processing, engineering drawing design and rendering, VDI is more exposed to the shortcomings of video redirection. In the process of playing high-definition video and graphics processing, there are some situations, such as flower screen and white screen feigning death. On the contrary, voi is more ideal in this aspect, which is no different from the operation of this machine. Therefore, according to the actual needs of the scene, it is more appropriate to choose the next generation cloud desktop with voi architecture.
5. Take line 8 from Wangjian station to zhaojiatiao station, transfer to line 3 to Huiji 2nd Road station, take exit b and walk to Wuhan municipal Party committee
6. Unknown_Error
7. According to my calculation, this dream journey to the West should cost 52000 yuan to increase its combat power by 20000 yuan.
8. Bitcoin does not belong to futures. Bitcoin is a kind of virtual currency, which is similar to stocks. Its price fluctuates sharply.
9. If the first charge in the early stage, it is recommended that the monkey king can use level 65 all the time. Other pets have no need. It's good to save some money to buy Yingzhao after level 65. If there is no money, it's good to be a vampire
10. Dream journey to the West 3D Putuo with Unicorn baby.
Hot content