websitedesign

Datacenter migration, when the network scale of data center becomes very large, it is necessary to add network equipment to realize multi-level connection.Present data centers tend to be a tree structure, the core put forward a few large capacity equipment, then hang under the multilayer devices (because the port number is not enough, may need to multilayer), dozens or even hundreds of sets of cascade network equipment together, once out of the fault, how can quickly find the fault equipment, often plagued by a lot of network operations staff.
The network equipment in the data center is redundant. As long as the fault device is found when the network fails, the business can be restored by isolating it, and then slowly troubleshoot the cause of the failure.Network failures usually get fault feedback from the application side first, and then start to troubleshoot. At this time, the application personnel usually only describe an application access failure phenomenon, and they will not tell you which specific addresses do not work to which addresses, and sometimes even wrong information, which greatly delays the problem positioning time.Problem location is spent most of the time in the troubleshooting process, how to do?How can data center networks quickly troubleshoot?This article will provide the answer.
If network failure is to be analyzed from the fault phenomenon of feedback from the application side, it is already too late, and it is easy for the application personnel to bring into the mistake area. Some of the feedback phenomenon of the application personnel is only seen by themselves, and the phenomenon is probably only a local phenomenon, which cannot reflect the failure situation of the entire network.Therefore, we should rely on ourselves to do a good job of network monitoring, and find problems through monitoring, so as to quickly find fault equipment, do equipment isolation or troubleshooting.
The early network monitoring was mainly to monitor some log and port traffic of the device. More often, this information was not enough and problems could not be detected in time.Many network device manufacturers say their device logs are complete, but there are still some extreme cases or software bugs that cause failure when there is no log output, so it is necessary to locate the traffic.At this time, it is necessary for the network personnel to find the application personnel to understand the fault phenomenon, find out some lost packets or blocked IP addresses on the spot, and then conduct network circulation. All the equipment that the fault traffic passes through will be circulated to find the fault equipment.Now that is a tree network, each layer there are a lot of equipment, the circulation is considerable, and not all of the devices are can do support of all the characteristics of the traffic statistics, there are not supported by the device will make statistics, increase the difficulty of finding fault equipment, do network operations over the years are so insist on. 
Obviously, the previous network troubleshooting method was effective, but the efficiency was too low, and the fault location time was long, which had a great impact on the business.Current network monitoring is aimed at data flow, monitoring specific data flow in the network, so as to find out the fault location once the data flow is interrupted.Here, several emerging network monitoring methods, also known as network visualization technology, are mentioned as the most effective methods for rapid obstacle elimination.
First of all, INT(in-band Network Telemetry, in-band Network Telemetry technology) technology enables the monitoring of Network state by collecting and reporting Network state at the data level.When the data message enters the first network equipment, equipment set on sampling and image out of the business flow of packet sampling way, INT based on packet encapsulation an INT head, and will need to collect information about the switch to fill in the INT data segment, message through all network equipment such processing, until finally a server connection of network equipment to spin out the INT head.
Message after each device will be collected INT message through gRPC message is sent to a remote monitoring server parsed and rendered, INT carried in the message delay of packet forwarding, equipment, congestion, etc., can be presented to the monitoring server, once the data message appear lost package or not, the monitoring server immediately perceived, and a few seconds to determine the scope of the problem and fault equipment.
Followed by ERSPAN (Encapsulated Remote Switch Port Analyzer, Remote monitoring of network traffic across the three layers of IP transmission technology), ERSPAN message based on the GRE encapsulation, and through the Ethernet forwarded to any IP routing can reach place.ERSPAN is to source port message a copy through the GRE (Generic Routing Encapsulation) sent to the server parsed, the physical location of collection server is not restricted, so that we can be the key to the whole network traffic forwarding through ERSPAN sent to the monitoring server, where traffic is part of the network appeared, be clear at a glance.The third is the sFlow and Netstream, these two are the data sampling technique, Netstream collection of relatively complete, but need to have dedicated hardware to complete, in the network deployment sFlow and Netstream, can pass gRPC will monitor data sent to the server, calculated by the monitoring server, and finishing, and graphical display the results appear, once part of the network has a problem, which can immediately appeared on the monitoring server.
SFlow and Netstream collect the main features of the header, rather than the whole content of the message. This is quite different from INT and ERSPAN, and there is no problem in troubleshooting most network faults.In a network, it does not mind the deployment of all three monitoring schemes, so that in case of failure, the data collected from multiple angles can be analyzed.
It is also important to try to send the data collection to the monitoring server through the management network. Otherwise, if there is a problem with the data network, the monitored data may not reach the monitoring server normally.In most cases, the failure of data network rarely affects the management network, and all devices can still be accessed normally. If the failure occurs, many devices cannot be accessed through the management network, it can be basically determined that this device is the fault point. 
With the above network monitoring methods, it is not difficult to find the fault at the first time, and it can be fully automated. When the fault is found, the monitoring server will automatically issue the isolation command to isolate the fault equipment and recover it automatically.In this way, the network fault location can be found before the application failure is reported, the fault equipment can be isolated in time, and the service can be restored. In this way, the time of fault analysis can be greatly shortened, which has little impact on the business, and even the business part cannot perceive the fault at all.The actual application effect of network monitoring technologies such as INT and ERSPAN is still unknown. These technologies are always mentioned recently and need to be tested in practice.SFLOW and Netstream are relatively mature technologies, but they are not really used in network troubleshooting, so they need to be popularized.
With these monitoring technologies, network failures can be quickly eliminated, which is of great significance to the operation and maintenance of data centers and greatly improves the operation and maintenance efficiency.

Datacenter migration, when the network scale of data center becomes very large, it is necessary to add network equipment to realize multi-level connection.Present data centers tend to be a tree structure, the core put forward a few large capacity equipment, then hang under the multilayer devices (because the port number is not enough, may need to multilayer), dozens or even hundreds of sets of cascade network equipment together, once out of the fault, how can quickly find the fault equipment, often plagued by a lot of network operations staff.
The network equipment in the data center is redundant. As long as the fault device is found when the network fails, the business can be restored by isolating it, and then slowly troubleshoot the cause of the failure.Network failures usually get fault feedback from the application side first, and then start to troubleshoot. At this time, the application personnel usually only describe an application access failure phenomenon, and they will not tell you which specific addresses do not work to which addresses, and sometimes even wrong information, which greatly delays the problem positioning time.Problem location is spent most of the time in the troubleshooting process, how to do?How can data center networks quickly troubleshoot?This article will provide the answer.
If network failure is to be analyzed from the fault phenomenon of feedback from the application side, it is already too late, and it is easy for the application personnel to bring into the mistake area. Some of the feedback phenomenon of the application personnel is only seen by themselves, and the phenomenon is probably only a local phenomenon, which cannot reflect the failure situation of the entire network.Therefore, we should rely on ourselves to do a good job of network monitoring, and find problems through monitoring, so as to quickly find fault equipment, do equipment isolation or troubleshooting.
The early network monitoring was mainly to monitor some log and port traffic of the device. More often, this information was not enough and problems could not be detected in time.Many network device manufacturers say their device logs are complete, but there are still some extreme cases or software bugs that cause failure when there is no log output, so it is necessary to locate the traffic.At this time, it is necessary for the network personnel to find the application personnel to understand the fault phenomenon, find out some lost packets or blocked IP addresses on the spot, and then conduct network circulation. All the equipment that the fault traffic passes through will be circulated to find the fault equipment.Now that is a tree network, each layer there are a lot of equipment, the circulation is considerable, and not all of the devices are can do support of all the characteristics of the traffic statistics, there are not supported by the device will make statistics, increase the difficulty of finding fault equipment, do network operations over the years are so insist on. 
Obviously, the previous network troubleshooting method was effective, but the efficiency was too low, and the fault location time was long, which had a great impact on the business.Current network monitoring is aimed at data flow, monitoring specific data flow in the network, so as to find out the fault location once the data flow is interrupted.Here, several emerging network monitoring methods, also known as network visualization technology, are mentioned as the most effective methods for rapid obstacle elimination.
First of all, INT(in-band Network Telemetry, in-band Network Telemetry technology) technology enables the monitoring of Network state by collecting and reporting Network state at the data level.When the data message enters the first network equipment, equipment set on sampling and image out of the business flow of packet sampling way, INT based on packet encapsulation an INT head, and will need to collect information about the switch to fill in the INT data segment, message through all network equipment such processing, until finally a server connection of network equipment to spin out the INT head.
Message after each device will be collected INT message through gRPC message is sent to a remote monitoring server parsed and rendered, INT carried in the message delay of packet forwarding, equipment, congestion, etc., can be presented to the monitoring server, once the data message appear lost package or not, the monitoring server immediately perceived, and a few seconds to determine the scope of the problem and fault equipment.
Followed by ERSPAN (Encapsulated Remote Switch Port Analyzer, Remote monitoring of network traffic across the three layers of IP transmission technology), ERSPAN message based on the GRE encapsulation, and through the Ethernet forwarded to any IP routing can reach place.ERSPAN is to source port message a copy through the GRE (Generic Routing Encapsulation) sent to the server parsed, the physical location of collection server is not restricted, so that we can be the key to the whole network traffic forwarding through ERSPAN sent to the monitoring server, where traffic is part of the network appeared, be clear at a glance.The third is the sFlow and Netstream, these two are the data sampling technique, Netstream collection of relatively complete, but need to have dedicated hardware to complete, in the network deployment sFlow and Netstream, can pass gRPC will monitor data sent to the server, calculated by the monitoring server, and finishing, and graphical display the results appear, once part of the network has a problem, which can immediately appeared on the monitoring server.
SFlow and Netstream collect the main features of the header, rather than the whole content of the message. This is quite different from INT and ERSPAN, and there is no problem in troubleshooting most network faults.In a network, it does not mind the deployment of all three monitoring schemes, so that in case of failure, the data collected from multiple angles can be analyzed.
It is also important to try to send the data collection to the monitoring server through the management network. Otherwise, if there is a problem with the data network, the monitored data may not reach the monitoring server normally.In most cases, the failure of data network rarely affects the management network, and all devices can still be accessed normally. If the failure occurs, many devices cannot be accessed through the management network, it can be basically determined that this device is the fault point. 
With the above network monitoring methods, it is not difficult to find the fault at the first time, and it can be fully automated. When the fault is found, the monitoring server will automatically issue the isolation command to isolate the fault equipment and recover it automatically.In this way, the network fault location can be found before the application failure is reported, the fault equipment can be isolated in time, and the service can be restored. In this way, the time of fault analysis can be greatly shortened, which has little impact on the business, and even the business part cannot perceive the fault at all.The actual application effect of network monitoring technologies such as INT and ERSPAN is still unknown. These technologies are always mentioned recently and need to be tested in practice.SFLOW and Netstream are relatively mature technologies, but they are not really used in network troubleshooting, so they need to be popularized.
With these monitoring technologies, network failures can be quickly eliminated, which is of great significance to the operation and maintenance of data centers and greatly improves the operation and maintenance efficiency.

Nowadays, website design is no longer strange to us, it is also a part of the development of enterprises. As a professional website design website, we all know that now we want to make the website better and get the user's approval, the details are particularly important. So, how can we do a good job of user experience? How can professional website design be able to impress users with details?
1. The arrangement of web pages
Professional website design on the web page arrangement, we need to create a simple, refined, high-end feeling, to know that the different websites in the arrangement, not enough attention to details, resulting in the final effect is not only very unprofessional, but will bring some disorderly feeling in the vision of the family. In the arrangement, the horizontal arrangement and the vertical arrangement are very important. The random arrangement is the big taboo of the website construction, because this can not reflect the sense of hierarchy, and will not create the concise and consistent visual effect.
2. The contrast effect of the color of the picture
In the course of professional website design, pictures and words are essential elements, color matching is particularly important, because to quickly let customers notice the key content of the site and receive the information of the web page in a short time, it is necessary to focus the attention of the visitors. Through color matching to achieve a visual contrast effect, is the most simple, direct, crude and effective way. Black and white is the most basic color matching problem, but also a simple style of the embodiment, different industries have different websites and elements, so we can adjust the collocation of these colors to form a more intense contrast effect.
3. The effect of the font
The effect of font effect on the construction of professional web pages is relatively large, because each type of font has a different flavor and style, and the sensory effects created are also different. In addition, the choice of fonts, such as size, thickness and color changes, need to be considered. Ensure that the font size of the page can be kept in a harmonious and unified state, which makes people seem effortless.
4. The homepage of the website
The first view of the user is the home page, which is the first impression that the user will see, the equivalent of the user's first impression of the enterprise website. It is self-evident. What kind of content do the users like? Most of the current users prefer simple, comfortable and simple pages, so when designing home pages in web design, Be sure to pay attention to the details and avoid carelessness in the details.
5. The framework is clear
Professional website design framework plays a guiding role in website content. When users are looking for requirements, this is an important part for us to display enterprise information. Therefore, in order to make customers more convenient and easy to find what they want to see, we should follow the framework of the website to provide users with an accurate page, so that users can realize the convenience of using our website.
6. At any time online communication
Users are confused or need service when they visit your web page. Our website must ensure that there is a special person to answer and communicate at any time. Our contact method is effective and allows users to find us in time. Do not miss any opportunity to communicate with customers. This is a very important link, and professional website design can not be ignored.
7. The overall coordination of the web site
Professional website design a web site overall coordination and aesthetic is very important, no matter how much our content, in the layout and classification we must avoid sloppy, messy. This is easy for users to grasp the main points, and do not know what to do. This detrimental to the overall aesthetic perception, thereby affecting the image of the enterprise.

Nowadays, website design is no longer strange to us, it is also a part of the development of enterprises. As a professional website design website, we all know that now we want to make the website better and get the user's approval, the details are particularly important. So, how can we do a good job of user experience? How can professional website design be able to impress users with details?
1. The arrangement of web pages
Professional website design on the web page arrangement, we need to create a simple, refined, high-end feeling, to know that the different websites in the arrangement, not enough attention to details, resulting in the final effect is not only very unprofessional, but will bring some disorderly feeling in the vision of the family. In the arrangement, the horizontal arrangement and the vertical arrangement are very important. The random arrangement is the big taboo of the website construction, because this can not reflect the sense of hierarchy, and will not create the concise and consistent visual effect.
2. The contrast effect of the color of the picture
In the course of professional website design, pictures and words are essential elements, color matching is particularly important, because to quickly let customers notice the key content of the site and receive the information of the web page in a short time, it is necessary to focus the attention of the visitors. Through color matching to achieve a visual contrast effect, is the most simple, direct, crude and effective way. Black and white is the most basic color matching problem, but also a simple style of the embodiment, different industries have different websites and elements, so we can adjust the collocation of these colors to form a more intense contrast effect.
3. The effect of the font
The effect of font effect on the construction of professional web pages is relatively large, because each type of font has a different flavor and style, and the sensory effects created are also different. In addition, the choice of fonts, such as size, thickness and color changes, need to be considered. Ensure that the font size of the page can be kept in a harmonious and unified state, which makes people seem effortless.
4. The homepage of the website
The first view of the user is the home page, which is the first impression that the user will see, the equivalent of the user's first impression of the enterprise website. It is self-evident. What kind of content do the users like? Most of the current users prefer simple, comfortable and simple pages, so when designing home pages in web design, Be sure to pay attention to the details and avoid carelessness in the details.
5. The framework is clear
Professional website design framework plays a guiding role in website content. When users are looking for requirements, this is an important part for us to display enterprise information. Therefore, in order to make customers more convenient and easy to find what they want to see, we should follow the framework of the website to provide users with an accurate page, so that users can realize the convenience of using our website.
6. At any time online communication
Users are confused or need service when they visit your web page. Our website must ensure that there is a special person to answer and communicate at any time. Our contact method is effective and allows users to find us in time. Do not miss any opportunity to communicate with customers. This is a very important link, and professional website design can not be ignored.
7. The overall coordination of the web site
Professional website design a web site overall coordination and aesthetic is very important, no matter how much our content, in the layout and classification we must avoid sloppy, messy. This is easy for users to grasp the main points, and do not know what to do. This detrimental to the overall aesthetic perception, thereby affecting the image of the enterprise.

機房建置，如何建設大型機房，建設一個數據中心機房不是一件簡單的事，數據中心的建設是一個整體結構，任何一個細節規劃忽略都有可能導致問題的發生。
甚至是設備或系統的一個生命週期內才能暴露出來，為避免數據中心建設完成交付後，因建設時未考慮周全出現的問題造成工期延誤而需要返工，以下是確保數據中心規劃設計人員堅持正確方向的關鍵問題。
一個全面的數據中心機房建設工程一般包括：綜合佈線、抗靜電地板鋪設、棚頂牆體裝修、隔斷裝修、ups電源、專用恆溫恆濕空調、機房環境及動力設備監控系統、新風系統、漏水檢測、地線系統、防雷系統、門禁、監控、消防、報警、屏蔽工程等。

機房建置，如何建設大型機房，建設一個數據中心機房不是一件簡單的事，數據中心的建設是一個整體結構，任何一個細節規劃忽略都有可能導致問題的發生。
甚至是設備或系統的一個生命週期內才能暴露出來，為避免數據中心建設完成交付後，因建設時未考慮周全出現的問題造成工期延誤而需要返工，以下是確保數據中心規劃設計人員堅持正確方向的關鍵問題。
一個全面的數據中心機房建設工程一般包括：綜合佈線、抗靜電地板鋪設、棚頂牆體裝修、隔斷裝修、ups電源、專用恆溫恆濕空調、機房環境及動力設備監控系統、新風系統、漏水檢測、地線系統、防雷系統、門禁、監控、消防、報警、屏蔽工程等。

Datacenter migration, liquid cooling solutions are expected to enter more enterprise data centers.In this paper, we will start from five aspects of the reasons and readers friends to explore.
Today, liquid cooling solutions that have traditionally been used primarily for mainframe and academic supercomputers may soon infiltrate more enterprise-class data centers.Now, as new and more demanding enterprise workloads continue to push up the power density of data center server racks, managers and operators of enterprise data centers are desperate to find alternatives that are more effective than air cooling systems.
We have interviewed a series of data center operators and suppliers and asked them for their views on the promotion of liquid cooling solutions to mainstream applications.Some of the respondents did not want to disclose the specific applications they used in their data centers and said they viewed the workloads and their cooling methods as a competitive advantage for their companies.
A series of super-sized cloud service operators, including Alphabet, parent companies of Microsoft, Google, Facebook and baidu, have formed a group dedicated to creating an open specification for liquid-cooled server racks, but the group has yet to specify what they will use.However, in these very large scale data centers, at least there is a certain type of workload obviously need to adopt liquid cooling scheme, i.e., by the GPU acceleration machine learning system (or for Google company, is the TPU tensor of the latest processor, which has said publicly that the TPU is now using direct cooling liquid cooling design of chip).
While current corporate data center operators are skeptical and concerned about the use of liquid cooling solutions, some trends are already emerging.If your enterprise supports any of the following workloads in the data center, your data center may also adopt liquid cooling scheme in the future:
1. AI and accelerator
The rate of annual CPU performance growth described by Moore's law has slowed sharply in recent years.This is partly because accelerators (mainly gpus), as well as FPGA and dedicated asics, are increasingly entering enterprise data centers.
Gpu-driven machine learning is probably the most common hardware acceleration use case outside of HPC(high-performance computing).However, in 451 by the market Research institutions Research recently conducted a survey, about a third of IT service provider, said their business plan in the online data mining, analysis and engineering simulation, real-time video and other media, fraud detection, load balancing, and a similar delay sensitive services using the acceleration in the system.
Hardware accelerators have much higher thermal design points (TDP, thermal design points) than cpus, and usually require 200W or more power to cool them.Add a high-performance server CPU, and a single system in your enterprise data center will require more than 1kW of power to cool it.
Intel is also aggressively breaking the 150W limit on its conventionally designed server processors."As more and more corporate customers want more powerful chips, we are starting to see a gradual increase in the amount of watt consumed by these chips.""Said Andy Lawrence, executive director of the Uptime Institute.
The rack density of enterprise data center servers is increasing.Most data centers now have at least some racks over 10kW in their normal operating orbits, while 20 percent of the racks even have 30kW or higher power densities.But these workloads are not considered high performance computing."They just say that their workload has a higher density frame.""Lawrence said.
"If you put the GPU together with Intel's processors, their power density could triple."He said.The liquid cooling scheme is obviously very suitable for these accelerators, especially the immersion cooling scheme, which can cool GPU and CPU.
2. Cooling high-density storage
As the current density of storage in enterprise data centers continues to increase, it may become more difficult to effectively cool storage.Most of the storage capacity installed in the data center is composed of unsealed hard disk drives and cannot be cooled by liquid cooling.Newer technologies, however, offer hope to business users in the industry.For example, solid state drives can be cooled using a full immersion solution.In addition, the creation of high-density, high-speed read/write head helium in the latest generation of storage hardware requires a sealed unit to make it suitable for liquid cooling schemes.
As noted in the 451 Research report, the combination of solid-state and helium-filled hard drives means there is no need to separate air-cooled storage from liquid-cooled processing.The increased reliability of hard drives also has a benefit: immersion in the drive in the coolant can help reduce the impact of heat and humidity on the components.
3. Network edge calculation
The need to reduce current and future application latency further drives the need for a new generation of data centers on the edge of the network.These can be high-density remote facilities deployed in wireless towers, factory operations workshops or retail stores.And these facilities may increasingly host high-density computing hardware, such as GPU packaging clusters for machine learning.
While not all edge data centers are liquid cooling solutions, many edge data centers will be designed to support heavy workloads in confined Spaces that cannot use traditional cooling solutions, or to cool in new deployment environments that do not use traditional preconditions.As a result of reduced energy consumption, liquid cooling makes it easier to deploy marginal sites where there is no large capacity for power.
As many as 20 percent of edge data centers can use liquid cooling, according to Lawrence's estimates.He envisions remote, micromodular, high-density data center sites supporting 40kW per rack.
4. High-frequency trading and blockchain
Many modern financial services firms are computationally intensive, requiring high performance cpus and gpus.These workloads include high frequency trading systems and blockchain based applications such as smart contracts and encrypted currencies.
For example, an enterprise client of GRC (Green Revolution Cooling), a high-frequency trading company, is testing its immersion Cooling solution.When green revolution cooling introduced immersive cooling products for the mining of encrypted currency and the price of bitcoin soared from the end of 2017, the company experienced its biggest ever sales surge.
Peter Poulin, GRC's chief executive, told reporters that another GRC corporate customer in Trinidad and Tobago is running an encrypted currency service at 100kW per rack and connecting a warm water cooling loop to the evaporating tower.Since warm water cooling is more energy efficient than cold water cooling, it can operate in a tropical environment without a mechanical cooler.
5. The cost of traditional cooling scheme is high
When air-based cooling systems cannot cope with high density cooling demands, liquid cooling schemes start to make sense.
Earth science company CGG, for example, using the GRC immersion liquid cooling system, in order to supply the data center in Houston cooling, CGG in the data center is mainly analyses of seismic data processing work, they use on commercial server is a powerful GPU, each frame up to 23 kw power.This power density is relatively high, but it is usually air-cooled.Ted Barragy, senior systems manager at the CGG, said: "we put heavy computing servers in immersion tanks to cool down.But the truth is that this is not so much about meeting the application's workload as about the cost economy of an immersive liquid cooling solution.
In its upgrade process, the immersion liquid cooling scheme replaces the traditional cooling equipment used by CGG's old data center.According to Barragy, the team restored several megawatts of power because of the upgrade."Even after a few years of adding servers and immersion tanks, we still have half a megawatt of unused power."He said."It's an old, traditional data center, and about half of its power is used for inefficient air cooling systems."
Barragy also said the PUE value of the submerged cooling data center was about 1.05.This is more efficient than the company's new but air-cooled data center in Houston, which has a PUE of 1.35.
"A lot of people think that liquid cooling is just a high-density cooling solution that is really suitable for calculating power density of 60kW to 100kW per rack, but it has other significant advantages for our mainstream corporate customers," Poulin said.
Chris Brown, chief technology officer at the Uptime Institute, said they have now seen a general increase in interest in liquid cooling solutions.This is driven by the urgent need of enterprise data centers to achieve higher efficiency and lower operating costs.
"The emphasis on liquid cooling is no longer on ultra-high density, but on solutions that can be used by the average enterprise data center operations manager to cool any IT assets."He said."The solution is now moving into more common density solutions and more common data centers."

Datacenter migration, liquid cooling solutions are expected to enter more enterprise data centers.In this paper, we will start from five aspects of the reasons and readers friends to explore.
Today, liquid cooling solutions that have traditionally been used primarily for mainframe and academic supercomputers may soon infiltrate more enterprise-class data centers.Now, as new and more demanding enterprise workloads continue to push up the power density of data center server racks, managers and operators of enterprise data centers are desperate to find alternatives that are more effective than air cooling systems.
We have interviewed a series of data center operators and suppliers and asked them for their views on the promotion of liquid cooling solutions to mainstream applications.Some of the respondents did not want to disclose the specific applications they used in their data centers and said they viewed the workloads and their cooling methods as a competitive advantage for their companies.
A series of super-sized cloud service operators, including Alphabet, parent companies of Microsoft, Google, Facebook and baidu, have formed a group dedicated to creating an open specification for liquid-cooled server racks, but the group has yet to specify what they will use.However, in these very large scale data centers, at least there is a certain type of workload obviously need to adopt liquid cooling scheme, i.e., by the GPU acceleration machine learning system (or for Google company, is the TPU tensor of the latest processor, which has said publicly that the TPU is now using direct cooling liquid cooling design of chip).
While current corporate data center operators are skeptical and concerned about the use of liquid cooling solutions, some trends are already emerging.If your enterprise supports any of the following workloads in the data center, your data center may also adopt liquid cooling scheme in the future:
1. AI and accelerator
The rate of annual CPU performance growth described by Moore's law has slowed sharply in recent years.This is partly because accelerators (mainly gpus), as well as FPGA and dedicated asics, are increasingly entering enterprise data centers.
Gpu-driven machine learning is probably the most common hardware acceleration use case outside of HPC(high-performance computing).However, in 451 by the market Research institutions Research recently conducted a survey, about a third of IT service provider, said their business plan in the online data mining, analysis and engineering simulation, real-time video and other media, fraud detection, load balancing, and a similar delay sensitive services using the acceleration in the system.
Hardware accelerators have much higher thermal design points (TDP, thermal design points) than cpus, and usually require 200W or more power to cool them.Add a high-performance server CPU, and a single system in your enterprise data center will require more than 1kW of power to cool it.
Intel is also aggressively breaking the 150W limit on its conventionally designed server processors."As more and more corporate customers want more powerful chips, we are starting to see a gradual increase in the amount of watt consumed by these chips.""Said Andy Lawrence, executive director of the Uptime Institute.
The rack density of enterprise data center servers is increasing.Most data centers now have at least some racks over 10kW in their normal operating orbits, while 20 percent of the racks even have 30kW or higher power densities.But these workloads are not considered high performance computing."They just say that their workload has a higher density frame.""Lawrence said.
"If you put the GPU together with Intel's processors, their power density could triple."He said.The liquid cooling scheme is obviously very suitable for these accelerators, especially the immersion cooling scheme, which can cool GPU and CPU.
2. Cooling high-density storage
As the current density of storage in enterprise data centers continues to increase, it may become more difficult to effectively cool storage.Most of the storage capacity installed in the data center is composed of unsealed hard disk drives and cannot be cooled by liquid cooling.Newer technologies, however, offer hope to business users in the industry.For example, solid state drives can be cooled using a full immersion solution.In addition, the creation of high-density, high-speed read/write head helium in the latest generation of storage hardware requires a sealed unit to make it suitable for liquid cooling schemes.
As noted in the 451 Research report, the combination of solid-state and helium-filled hard drives means there is no need to separate air-cooled storage from liquid-cooled processing.The increased reliability of hard drives also has a benefit: immersion in the drive in the coolant can help reduce the impact of heat and humidity on the components.
3. Network edge calculation
The need to reduce current and future application latency further drives the need for a new generation of data centers on the edge of the network.These can be high-density remote facilities deployed in wireless towers, factory operations workshops or retail stores.And these facilities may increasingly host high-density computing hardware, such as GPU packaging clusters for machine learning.
While not all edge data centers are liquid cooling solutions, many edge data centers will be designed to support heavy workloads in confined Spaces that cannot use traditional cooling solutions, or to cool in new deployment environments that do not use traditional preconditions.As a result of reduced energy consumption, liquid cooling makes it easier to deploy marginal sites where there is no large capacity for power.
As many as 20 percent of edge data centers can use liquid cooling, according to Lawrence's estimates.He envisions remote, micromodular, high-density data center sites supporting 40kW per rack.
4. High-frequency trading and blockchain
Many modern financial services firms are computationally intensive, requiring high performance cpus and gpus.These workloads include high frequency trading systems and blockchain based applications such as smart contracts and encrypted currencies.
For example, an enterprise client of GRC (Green Revolution Cooling), a high-frequency trading company, is testing its immersion Cooling solution.When green revolution cooling introduced immersive cooling products for the mining of encrypted currency and the price of bitcoin soared from the end of 2017, the company experienced its biggest ever sales surge.
Peter Poulin, GRC's chief executive, told reporters that another GRC corporate customer in Trinidad and Tobago is running an encrypted currency service at 100kW per rack and connecting a warm water cooling loop to the evaporating tower.Since warm water cooling is more energy efficient than cold water cooling, it can operate in a tropical environment without a mechanical cooler.
5. The cost of traditional cooling scheme is high
When air-based cooling systems cannot cope with high density cooling demands, liquid cooling schemes start to make sense.
Earth science company CGG, for example, using the GRC immersion liquid cooling system, in order to supply the data center in Houston cooling, CGG in the data center is mainly analyses of seismic data processing work, they use on commercial server is a powerful GPU, each frame up to 23 kw power.This power density is relatively high, but it is usually air-cooled.Ted Barragy, senior systems manager at the CGG, said: "we put heavy computing servers in immersion tanks to cool down.But the truth is that this is not so much about meeting the application's workload as about the cost economy of an immersive liquid cooling solution.
In its upgrade process, the immersion liquid cooling scheme replaces the traditional cooling equipment used by CGG's old data center.According to Barragy, the team restored several megawatts of power because of the upgrade."Even after a few years of adding servers and immersion tanks, we still have half a megawatt of unused power."He said."It's an old, traditional data center, and about half of its power is used for inefficient air cooling systems."
Barragy also said the PUE value of the submerged cooling data center was about 1.05.This is more efficient than the company's new but air-cooled data center in Houston, which has a PUE of 1.35.
"A lot of people think that liquid cooling is just a high-density cooling solution that is really suitable for calculating power density of 60kW to 100kW per rack, but it has other significant advantages for our mainstream corporate customers," Poulin said.
Chris Brown, chief technology officer at the Uptime Institute, said they have now seen a general increase in interest in liquid cooling solutions.This is driven by the urgent need of enterprise data centers to achieve higher efficiency and lower operating costs.
"The emphasis on liquid cooling is no longer on ultra-high density, but on solutions that can be used by the average enterprise data center operations manager to cool any IT assets."He said."The solution is now moving into more common density solutions and more common data centers."

Enterprise website design revision will cause website ranking and volatility, serious or even a big decline. So we must consider a clear and comprehensive analysis of the needs of the website before starting the website design, making it clear how to change the website to be more favorable. And what is the main content that needs to be restructured. If you haven't considered it clearly, I suggest that you do not alter it easily.
What are the problems that need to be paid attention to when designing and revising websites?
1, the reversion and uploading should be done in one time
When we are revising the design of enterprise website, we must fix it at once, do not change it a bit, otherwise it will be very unfavorable to the optimization of website. Want to do a sex to fix the problem, you must make sure that the website wants to change what it looks like, and do it with other pages, and then it can be modified slowly, not to change a little bit, because if the web site uploads new content, a search engine spiders grab the trouble. . This will help improve the trust of the enterprise website.
2. Determine the time of reversion
When the website design is reformed, it is necessary to determine the time for the reversion of the website. It is best to make the reversion when the search engine is the least active spider, so the reorganization of the enterprise website is not easy to be caught by the spider, and what problems in the process of the website reversion can be quickly modified, so the maximum reduction is also reduced. Less the impact of the web site.
3, the path of enterprise website revision must not be changed
In the design of the enterprise website, do not modify the path of the web page, because if the previous pages were included, the change of the path will affect the site, so that the original link is dead chain, can not be opened. This will cause the website to be friendly to search engines. For example, the pages that have been included before, are included in the path, do not change, or because of the change of the path, leading to the search engine to reduce the friendliness of the enterprise website.
4. Do not modify the keyword layout of the website content
When the website design is reformed, we must keep the key words of the content. If there is no change in the key words that are included and ranked, we can keep it as far as possible so as to minimize the impact of the reversion.
What are the main contents of the website construction reversion?
1. Function upgrade
Websites that are rarely visited by search engines are mostly websites with simple functions and poor information. In order to make browsing users and network management easy to operate and to better search engine, it is very important to add some relatively urgent functions in website construction.
2. Structural reorganization
In fact, many websites are just a picture of deceiving the naked eye built by the website design company in order to mix up the enterprises. Therefore, website design and revision should have a better search and ranking for your website. The website structure can not be ignored.
3. Establish navigation
The navigation of the website design reversion is not only for the browsing users, but also for the search robot. It must let the search engine know what kind of pages we want it to update frequently, and what kind of information he wants to recommend. However, almost 95% of Chinese websites do not have navigation.
4. Data optimization
The websites we see by the naked eye are actually derived from source code. Too long the source code will not only affect the opening speed of the website, but also affect the ranking. In particular, the visual effects generated by JS code are very much disliked by search engines.

Enterprise website design revision will cause website ranking and volatility, serious or even a big decline. So we must consider a clear and comprehensive analysis of the needs of the website before starting the website design, making it clear how to change the website to be more favorable. And what is the main content that needs to be restructured. If you haven't considered it clearly, I suggest that you do not alter it easily.
What are the problems that need to be paid attention to when designing and revising websites?
1, the reversion and uploading should be done in one time
When we are revising the design of enterprise website, we must fix it at once, do not change it a bit, otherwise it will be very unfavorable to the optimization of website. Want to do a sex to fix the problem, you must make sure that the website wants to change what it looks like, and do it with other pages, and then it can be modified slowly, not to change a little bit, because if the web site uploads new content, a search engine spiders grab the trouble. . This will help improve the trust of the enterprise website.
2. Determine the time of reversion
When the website design is reformed, it is necessary to determine the time for the reversion of the website. It is best to make the reversion when the search engine is the least active spider, so the reorganization of the enterprise website is not easy to be caught by the spider, and what problems in the process of the website reversion can be quickly modified, so the maximum reduction is also reduced. Less the impact of the web site.
3, the path of enterprise website revision must not be changed
In the design of the enterprise website, do not modify the path of the web page, because if the previous pages were included, the change of the path will affect the site, so that the original link is dead chain, can not be opened. This will cause the website to be friendly to search engines. For example, the pages that have been included before, are included in the path, do not change, or because of the change of the path, leading to the search engine to reduce the friendliness of the enterprise website.
4. Do not modify the keyword layout of the website content
When the website design is reformed, we must keep the key words of the content. If there is no change in the key words that are included and ranked, we can keep it as far as possible so as to minimize the impact of the reversion.
What are the main contents of the website construction reversion?
1. Function upgrade
Websites that are rarely visited by search engines are mostly websites with simple functions and poor information. In order to make browsing users and network management easy to operate and to better search engine, it is very important to add some relatively urgent functions in website construction.
2. Structural reorganization
In fact, many websites are just a picture of deceiving the naked eye built by the website design company in order to mix up the enterprises. Therefore, website design and revision should have a better search and ranking for your website. The website structure can not be ignored.
3. Establish navigation
The navigation of the website design reversion is not only for the browsing users, but also for the search robot. It must let the search engine know what kind of pages we want it to update frequently, and what kind of information he wants to recommend. However, almost 95% of Chinese websites do not have navigation.
4. Data optimization
The websites we see by the naked eye are actually derived from source code. Too long the source code will not only affect the opening speed of the website, but also affect the ranking. In particular, the visual effects generated by JS code are very much disliked by search engines.