The function of the data center expert management CRAC

CRAC has developed a number of other positions, such as cloud engineers, IT architects, and other positions with the rapid development of technology in the data center. But data center experts are still the mainstay of the day-to-day operation of the data center, and there are some important functions to manage the data center in the IT organization.

Monitoring system: Data Center experts implement, support and use various monitoring and management tools in application programs, resource repositories or physical facilities, observe critical alerts, and respond to events in time. They can use real-time monitoring to observe the operation process of each system and improve progress, such as allocate more storage space to the current storage restricted workload.
Integration: Data Center experts need to ensure that systems, services, and applications can work properly in deployment or integration. This requires an in-depth understanding of the system configuration and the interdependence of the system application components within the data center. Experts need to constantly improve or upgrade some components, such as installing system, maintenance system, cable and other infrastructure constantly, so as to save money and improve performance.
Troubleshooting: Data Center experts need to access logs and monitoring data, provide first level and two level support, and use root cause analysis technology to troubleshoot. Experts follow the established event management procedures to ensure that the IT Department has fully responded, and fully communicated with the downtime problem, ensuring real-time follow-up and timely and proper solution of the problem. Experts can also take positive measures according to their own experience, so as to reduce or effectively prevent problems again.
Collaboration: Data Center experts and users and other IT members need to interact closely, which requires clear written and oral communication, and often leads to articles, guidebooks and other contents for IT staff and users. The development of agile software development paradigms, such as DevOps, also emphasizes the need for collaboration, and builds operational support for a continuous software development and release cycle.

機房建置如何減少人爲故障發生率

機房建置企業常常因爲運維管理人員的操作不當問題而出現硬件和網絡故障等問題。
那麼不管是在機房還是遠程運維的工作人員選擇何種日常事務處理方式來高效安全工作？

（1）明確穩健的流程和文檔
在數據中心進行的操作過程都應該文檔化，有着明確具體的驗證和實踐過得程序來進行。
當然在開始的時候需要數據中心管理人員花費時間和精力來創建、記錄與維護這些流程和程序，建立程序庫並對工作員工進行培訓和學習，可以有效避免因爲操作不當引起的網絡問題。
（2）上崗前進行專業知識培訓
數據中心工作人員應該瞭解電氣和機械系統的基本知識，數據中心繫統之間的相互關係，以及如何解決在這些類型的環境中可能出現的常見問題。
此外，工作人員還應具有良好的解釋能力和分析解決問題的能力。
爲了建立一致的基礎知識，服務供應商也應該定期培訓他們的員工。
McClary指出，許多數據中心設施運營商只提供短暫的工作培訓，但不一定會長期進行。
培訓必須持續開展，而每個員工都應該對自己的教育和能力負責。
記錄的流程和程序可爲培訓工作奠定基礎。
隨着知識範圍的不斷變化和擴展，額外的培訓可以確保對每個工作人員的角色，責任，以及所需技能有着敏銳的瞭解。
（3）日常檢查和演練
數據中心員工花費時間去體驗並檢查數據中心設施中的所有關鍵系統至關重要。
這些演練可以與培訓工作結合起來，幫助工作人員認識到關鍵組成部分和任何可能出現的問題。
數據中心管理人員應該通過他們的檢查來制定一些文檔化的程序來幫助指導這些工作。
這包括在演練期間應該檢查的項目的列表，工作人員應該記錄的具體參數，以及在參數結果中應採取的步驟。
通過演練可以幫助工作人員找出容易糾正的問題，防止以後出現更大的問題。
數據中心在進行租機服務時，通過手動操作來機房佈線、上架服務器、安裝系統、分配IP、添加硬盤等，不可避免的出現一些誤操作，用戶在遇到這類問題時可以督促運維工作人員仔細，同時也可以適當的理解這種錯誤的出現。
現在比較先進的鏡像和備份功能，對於數據丟失問題有一定的解決作用。
總而言之，在完美的設備沒有完美的管理措施是容易發生事故的。
數據中心的所有管理人員只有熟悉自己是誰，自己要做什麼才能真正確保數據中心安全的運轉。

Data center migration: how to store data?

Data center migration: choosing a data center outside a big city like London is a good innovation case. In fact, this big city has been selected historically. Although a city like London seems to be the best choice for business projects, it is important to understand that this is not the only option, because the rise of communities like Ireland has proved this.

Location plays an important role in our decision-making process. From where to choose to live, where to work, where to go on vacation, even where to do business, and how we manage our information.
The growth of data volume plus the demand for more flexible and customized IT capabilities means that a lot of money is needed to manage and maintain its internal IT infrastructure, which is becoming more and more difficult to prove. On the contrary, the third party online stores, such as hosting facilities, and the flexibility provided by these providers now means that you can transfer responsibilities to others, making you feel at ease, knowing that your IT infrastructure is secure. But once you decide to outsource, how will you choose a provider?
Historically, London is usually the best place to store information, but with the consideration of factors such as rent, scalability and data growth, operators began to have more consideration, so there are more locations to choose. Recently, a report from BroadGroup, a data consultancy, shows that Ireland is the best place to build facilities in Europe. The reason is the integration of connectivity among cities, preferential taxation and active government support. Among them, both Amazon and Microsoft have offices in Dublin, and it is worth mentioning that this is one of Microsoft's largest offices in Europe.
Now, apple expects to build a data center worth 850 million euros in Athenry outside Dublin. This behavior has led more and more enterprises to consider the location of traditional data centers. So what are the factors to be considered when looking for a position?
First, the facilities in the main cities will automatically involve the costs and risks associated with urban life. In places like London, where real estate is rare, the rental market is very competitive, so the cost is transferred to the customer.
Another consideration is the future cost - whether your business is likely to grow, which means you may need more rack space. If this hypothesis is just set up, it is necessary to ensure that the center can provide space for growth, because it is expensive to move the facilities to a new environment. The center outside London is often larger, with greater development and size, so it is easy to integrate into new technology.
While the location related costs may be specific to a particular business, security and risk are a common concern in the industry. In the traditional sense, data centers are often close and connected to each other. In London, you will find that most of the data centers are located in the east of the city. Though there are obvious convenience conditions near financial area and important exchanges, there are potential risks at the same time, especially putting all data centers in one basket.
People live in the real society of terrorist threats. Recent attacks in London and Manchester are a serious reminder, and they also indicate intermittently that we cannot predict the coming of another event. A city that attacks will probably enter the blockade. The measures of remote guidance can reduce the loss of this problem. But what will happen if your data center has problems and no one will solve this problem?
In addition, the natural environment of the data center is also an important element of the definition of security, which essentially increases or reduces enterprise risk. The data center is located near the river, such as Thames River, which directly affects the operation of the data center in the case of flooding. Locating a data center in the flooded area has always been a dangerous strategy. What we need to protect is that if the flood prevention system fails, the whole IT system will be destroyed.
For large cities, fire is also a persistent threat. Though today's advanced fire protection system has developed a lot in protecting urban data centers, it still has great risk compared with those outside the city's data centers. These are things that enterprises should consider when looking for data center partners. For each company, each business has its own unique needs, and its facilities are very important to meet these needs. Even if you want a big city like London, the outsourced information seems to be the choice of immortals, but it's important that this is not the only choice. Because of the rise of Ireland and other areas, this has been confirmed.  Recognizing this point will allow you to gain more features and benefits, reduce the threat of risk, and reduce the cost significantly.

Website design: what do you need to pay attention to in setting up a backstage system?

Low coupling of website design function modules
According to the business process, some basic functional modules are derived from business management, commodity management, order management, logistics after-sale, payment settlement and account management. Each module also contains multiple menu functions.

Business management: include business registration landing, company information management, qualification certification, pay deposit. Commodity management: manage commodity brand category, publish commodity, commodity shelf, commodity shelf, commodity warehouse. Order management: order view, abnormal order processing, and exchange document management.
Logistics after sale: logistics distribution setting, delivery management.
Payment settlement: settlement management, settlement details, reimbursement documents management, freight settlement management.
Account management: margin system, goods payment inquiry, account amount inquiries.
With the specific module, we can refine the specific menu function. With the growth of business and scale growth, some functional modules will be stripped to separate systems. Therefore, when designing functional modules, we need to pay attention to low coupling and reconfiguration day, so that we can migrate better. Many of the background products need to understand and even understand the technology, because the background system design, we need to refine each field, understand the database, table structure. Three. The flow of data is blood
A product has two kinds of data: input data, the flow of data refers to the direction of the input data of the system, the source of the output data. The flow of data makes the product alive.
The background system is not a single existence, a merchant system interacts with the investment management system, product system, order system, payment system, financial system, logistics system, BI system, marketing system, advertising system, recommendation system, message center...
In product design, we need to know clearly that every function module interacts with the system, and the whereabouts and circulation of each single data involve the influence of related systems. In this way, the risk can be avoided. At the same time, the extensibility and extensibility of the controlled products are also extended.
The following points need to be paid attention to in the design of the backstage system.
1, the system is easy to understand.
A good and easy to use system can greatly improve the efficiency of the related staff and colleagues, and directly save time and cost.
2, the function module is low coupling.
Designing products that make the architecture of Internet products is growing very fast, iterative frequent, user growth is tens of thousands, maybe 35 months, the existing system has not meet the current business, and module division, good low coupling, whether in the iterative reconstruction, or data migration, will play great role.  3. Distribution of authority roles.
The background system is a data pool, the related workflow is assisted by a variety of roles or many departments, so the relevant privileged roles need to be well divided.
First create a role - and then assign permissions to the role, different roles, menus, and data permissions.
Create an account again - assign roles to the account. In this way, the user allocates the functional permissions.

The FM200 Internet of things is changing the demand for data centers

FM200：At present, modern organizations are using technologies that are higher and more advanced than ever before to respond to changing markets.

These companies are constantly making innovations, creating new business models, offering new products, new approaches, and new market competition to meet the challenges of competitors.
The data center, one of the largest and the most significant change is - from the true sense, driven by the end user needs, IT spending, mobility, data transmission has changed the modern user consumption habits.
More importantly, the new concept of the Internet of things (IoT) will have a significant impact on how data centers operate.
Here's a very interesting idea - a very cool topic that has been mentioned in recent conversations with partners - cloud connection recycling stations.
It's not just a recycle bin, it's more like a warehouse.
Here, creating an efficient waste management system has had little direct impact on the company's bottom line.
The warehouse knows which is complete, which can change the routing in real time, and which places need to be cleared.
All of a sudden, the "things" that the cloud never faced were improving business efficiency at the fastest rate.
There are other improvements, of course.
For example, tesla already supports HTML5 in its central system, which will soon be extended to more connected iot endpoints.
The current trend of data mobilization and growth indicates that the Internet of things is a booming concept.
(note: HTML5: the fifth major revision of an application of HTML), the core language of the world wide web, and an application in the standard generic markup language (HTML).
On October 29, 2014, the world wide web consortium announced that after nearly eight years of hard work, the standard specification was finally completed.
Browsers that support Html5 include Firefox (Firefox), IE9 and its higher version, Chrome (Google browser), Safari, Opera, etc.
Domestic Maxthon (Maxthon), and based on IE or Chromium (Chrome project or the experimental edition) 360 launched by the browser, sogou browser, QQ browser, such as cheetahs browser domestic browser also have the ability to support HTML 5.)
Cisco Visual Networking index report shows as follows: the smarter end user devices and M2M connection is the significant growth of clear indicators to the growth of the Internet of things, including people, processes, data and the "thing" together, make the network interconnection is more relevant and more valuable.
M2M and wearable devices are making computing and connectivity in our daily lives very popular.
(note: M2M is a concept and a general term for all technologies that enhance the communication and network capabilities of machine devices.
Communication between people is also achieved through machines, such as communication between people through mobile phones, telephones, computers and fax machines.
Another kind of technology is specially designed for machines and establish communication, such as many intelligent instrument with RS - 232 interface and GPIB communication interface, strengthened between instrument and apparatus, communication between the instrument and computer skills.
With the development of science and technology, more and more devices have the ability of communication and networking, and Network Everything has gradually become a reality.
Communication between people need to be more intuitive, beautiful interface, and richer multimedia content, and M2M communication more need to establish a unified, standardized communication interface and standardized content transmission.).
One of the most important reasons for the growing adoption of the Internet of things is the emergence of wearable electronics, a force with high growth potential.
As the name implies, wearable equipment is to be worn by people, and can be embedded cellular connection or through the use of wi-fi, bluetooth, bluetooth and other equipment (mainly smartphones) the technique of direct connection and communication to the network.
How does all this end up affecting data center requirements?
What should data center managers pay attention to when creating a next-generation data center platform?
Finally, how does the Internet of things reshape the framework of the data center?
When it comes to the overall planning and design of data centers, IoT actually has a major impact on the future planning of data centers.
Everything from rack density to DCIM will be affected by IOT.
Let's take a look at several aspects of the Internet of things that will change.
The number of new devices connected to data centers will continue to increase.
In addition, all these new connections are creating energy and resource utilization.
Buildings and their related processes (the U.S. energy information administration, eia. Doe. Gov) and hvac systems consume about 40 percent of their energy consumption, which may be an important factor in energy use.
IoT will create a number of new custom connections into modern data centers.
With this in mind - it makes sense to use custom cooling and intelligent air flow management solutions.
It is important to find partners and suppliers to help you meet the cooling requirements for energy efficiency goals.
There are powerful solutions that use the latest technologies to develop and integrate components into the cooling system to provide you with the level of efficiency required for the iot project.
Air quality and filtering: let's say you're in the hospital with IOT. They have multiple IOT devices connected to processing medical information, collaborating with patients, and even helping the operating room.
All of these devices are connected to the data center.
Now we are introducing more devices into the "cleaner" part of the organization - the air that optimizes the breathing of the infrastructure is becoming critical.
Consider this - the epa and Lawrence.
The Berkeley national laboratory estimates that we spend as much as 87 percent of the cost of indoor environments to enhance the importance of a healthy indoor environment.
Data center design leader is to provide a powerful clean room technology, can get rid of the nano-scale particles, which means that no matter whether you through a MERV 13 filter or develop clean room to meet the requirements of LEED credit, have technology can achieve.
Regarding the Internet of things, we need to find products that can handle many applications of indoor air quality, such as hospitals, schools and museums.
As the Internet of things continues to emerge, the use of data center resources will continue to grow.
This means that all aspects of data center resource management need to be managed as efficiently as possible.
One way to create a data center method that is easy to modify, upgrade, and maintain.
For example, the size of the new modular air processor is applicable to existing installation dimensions, corridor access, elevator, elevator and other access points.
In addition, the appropriate configuration of airflow management solutions helps the data center's airflow organization to be more reasonable.
These measures can help the data center to respond quickly to changing demands of users and the need for their own resources.
It will be a few years before we are fully connected to the Internet of things, and it is just beginning.
More and more devices and products are being connected in our lives -- radios in the car, when a computer dustbin is full, will prompt you to be full, and your fridge is full of milk and eggs.
In fact, all of this will permeate the need for data centers, resource utilization, higher user density levels, and so on.
Over the next few years, our world will be more interconnected.
Therefore, make sure your data center is ready for this.

機房建置裝修標準規範要求有哪些？

機房建置規範總要求
● 計算機房的室內裝修工程施工驗收主要包括吊頂、隔斷牆、門、窗、牆壁裝修、地面、活動地板的施工驗收及其他室內作業。

● 室內裝修作業應符合《裝飾工程施工及驗收規範》、《地面及樓面工程施工及驗收規範》、《木結構工程施工及驗收規範》及《鋼結構工程施工及驗收規範》的有關規定。
● 在施工時應保證現場、材料和設備的清潔。
隱蔽工程（如地板下、吊頂上、假牆、夾層內）在封口前必須先除塵、清潔處理，暗處表層應能保持長期不起塵、不起皮和不龜裂。
● 機房所有管線穿牆處的裁口必須做防塵處理，然後對縫隙必須用密封材料填堵。
在裱糊、粘接貼面及進行其他塗復施工時，其環境條件應符合材料說明書的規定。
● 裝修材料應儘量選擇無毒、無刺激性的材料，儘量選擇難燃、阻燃材料，否則應儘可能塗防火塗料。
機房裝修標準規範要求有哪些？
室內裝修
1、吊頂
● 計算機機房吊頂板表面應平整，不得起塵、變色和腐蝕；
其邊緣應整齊、無翹曲，封邊處理後不得脫膠；
填充頂棚的保溫、隔音材料應平整、乾燥，並做包縫處理。
● 按設計及安裝位置嚴格放線。
吊頂及馬道應堅固、平直，並有可靠的防鏽塗復。
金屬連接件、鉚固件除鏽後，應塗兩遍防鏽漆。
● 吊頂上的燈具、各種風口、火災探測器底座及滅火噴嘴等應定準位置，整齊劃一，並與龍骨和吊頂緊密配合安裝。
從表面看應佈局合理、美觀、不顯凌亂。
● 吊頂內空調作爲靜壓箱時，其內表面應按設計要求做防塵處理，不得起皮和龜裂。
● 固定式吊頂的頂板應與龍骨垂直安裝。
雙層頂板的接縫不得落在同一根龍骨上。
● 用自攻螺釘固定吊頂板，不得損壞板面。
當設計未作明確規定時應符合五類要求。
● 螺釘間距：沿板周邊間距150-200mm，中間間距爲200-3000mm，均勻佈置。
● 螺釘距板邊10-15mm，釘眼、接縫和陰陽角處必須根據頂板材質用相應的材料嵌平、磨光。
● 保溫吊頂的檢修蓋板應用與保溫吊頂相同的材料製作。
● 活動式頂板的安裝必須牢固、下表面平整、接縫緊密平直、靠牆、柱處按實際尺寸裁板鑲補。
根據頂板材質作相應的封邊處理。
● 安裝過程中隨時擦拭頂板表面，並及時清楚頂板內的餘料和雜物，做到上不留餘物，下不留污跡。
2、隔斷牆
● 無框玻璃隔斷，應採用槽鋼、全鋼結構框架。
牆面玻璃厚度不小於10mm，門玻璃厚度不小於12mm.表面不鏽鋼厚度應保證壓延成型後平如鏡面，不不平的視覺效果。
● 石膏板、吸音板等隔斷牆的沿地、沿頂及沿牆龍骨建築圍護結構內表面之間應襯墊彈性密封材料後固定。
當設計無明確規定時固定點間距不宜大於800mm.
● 豎龍骨準確定位並校正垂直後與沿地、沿頂龍骨可靠固定。
● 有耐火極限要求的隔斷牆豎龍骨的長度應比隔斷牆的實際高度短30mm，上、下分別形成15mm膨脹縫，其間用難燃彈性材料填實。
全鋼防火大玻璃隔斷，鋼管架刷防火漆，玻璃厚度不小於12mm，無氣泡。
● 安裝隔斷牆板時，板邊與建築牆面間隙應用嵌縫材料可靠密封。
● 當設計無明確規定時，用自攻螺釘固定牆板宜符合：螺釘間距沿板周邊間距不大於200mm，板中部間距不大於300mm，均勻佈置，其他要求同2.
●有耐火極限要求的隔斷牆板應與豎龍骨平等鋪設，不得與沿地、沿頂龍骨固定。
● 隔斷牆兩面牆板接縫不得在同一根龍骨上，每面的雙層牆板接縫亦不得在同一根龍骨上。
● 安裝在隔斷牆上的設備和電氣裝置固定在龍骨上。
牆板不得受力。
● 隔斷牆上需安裝門窗時，門框、窗框應固定在龍骨上，並按設計要求對其縫隙進行密封。
3、鋁合金門窗和隔斷
● 鋁合金門框、窗框、隔斷牆的規格型號應符合設計要求，安裝應牢固、平整，其間隙用非腐蝕性材料密封。
當設計無明確規定時隔斷牆沿牆立柱固定點間距不宜大於800mm.
● 門扇、窗扇應平整、接縫嚴密、安裝牢固、開閉自如、推拉靈活。
● 施工過程中對鋁合金門窗及隔斷牆的裝飾面應採取保護措施。
● 安裝玻璃的槽口應清潔，下槽口應補墊軟性材料。
玻璃與扣條之間按設計要求填塞彈性密封材料，應牢固嚴密。
4、活動地板
● 計算機房用活動地板應符合國標GB6650-86《計算機房用活動地板技術條件》。
● 活動地板的理想高度在18-24英寸（46-61cm）之間。
● 活動地板的鋪設應在機房內各類裝修施工及固定設施安裝完成並對地面清潔處理後進行。
● 建築地面應符合設計要求，並應清潔、乾燥，活動地板空間作爲靜壓箱時，四壁及地面均就作防塵處理，不得起皮和龜裂。
● 現場切割的地板，周邊應光滑、無毛刺，並按原產品的技術要求作相應處理。
● 活動地板鋪設前應按標高及地板佈置嚴格放線將支撐部件調整至設計高度，平整、牢固。
● 活動地板鋪設過程中應隨時調整水平。
遇到障礙或不規則地面，應按實際尺寸鑲補並附加支撐部件。
● 在活動地板上搬運、安裝設備時應對地板表面採取防護措施。
鋪設完成後，做好防靜電接地。

How enterprises reduce the cost of data center migration

Data center migration: the measurement of large data may be daunting for most people, but people in the IT department know how to deal with it. With the exponential growth of the functions of modern data centers, measurement indicators and key performance indicators (KPI) become more and more important for insight into all data. The most important indicator is the bottom line of the business.

This is not only a matter of measurement, but a management problem. It may be profitable to provide SaaS products, but like any business, there are some flaws that need to be noticed. Businesses need to "organically" develop their data centers. If you know where to reduce costs and increase the structure, business will be successful.
Of course, the use of the Internet can also bring some harm to enterprise operation. For example, 6% of the world's computers suffer from loss of data every year, but the benefits of interconnected business are too big to be abandoned. If an enterprise ensures active monitoring of its own data center, it can not only ensure the security of the network, but also ensure the cost-effectiveness.
Understand the overall cost of ownership (TCO)
The overall cost of ownership (TCO) is not a new concept, and no high-level definition is needed. The overall cost of ownership is this, but in depth some details are found to be taken into consideration.
Fast overall cost of ownership (TCO) calculation means considering two types of costs: direct and indirect costs.
Direct costs come from assets, such as enterprises' cars, printers, electricity, network connections, and wages.
The indirect cost is the cost that is indirectly related to the final goal. For example, the rent of the office space, the loss of productivity, the cost of the law.
If the enterprise can accurately calculate direct and indirect costs, the overall cost of ownership (TCO) can be determined. Now, how do enterprises determine the high cost, how to run the data center more effectively, and gain more profits?
TCO:IT Brand Pulse in the real world
IT Brand Pulse (ITBP), a research website providing data center architecture, data and analysis, released a case study in 2015, which investigated the cost of several components that owned and operated mass storage arrays. It takes into account the purchase price of multiple hypothetical hardware components of the data center, as well as two costs such as spare parts, training and annual support costs.
After explaining the methods used in the report, ITBP explored the available storage options. It weighs the performance of different solutions, including Amazon's cloud storage services, multiple disk arrays, and server based software definition solutions.
Concluding remarks
Each solution is different, but some basic concepts are applicable to all technology enterprises. They also apply to keeping low TCO for enterprise data centers:
• the end user is always considered first. Choosing a system with less function, simpler use and lower running costs may be more profitable, although it is known to do more. The features that seem to be easy to use for the experts in a team may not be suitable for the public, and increase the cost and complexity.
• decompose the deployment and ongoing enhancement into manageable projects.  The direct and indirect costs of each record, capital expenditure and operating cost are predicted and tracked. This will give the company an accurate understanding of the expenditure for a period of time, and enable enterprises to understand the location of extra cost or generate extra funds.
Communicate with employees. Keeping TCO means making changes. People may not respond to these changes in the way they want. This is inevitable. But employees will react more favorably to their knowledge of upcoming changes, let them know what will happen, and they will put forward valuable opinions and remain loyal.
Today, enterprise owners / operators can use tools and knowledge to grant them access to more critical data.

What are the key points of the website design?

Now the enterprises are more important the website design, but for the network industry, it is not clear to these. We can only watch the surface phenomenon, but we can't see the deep content.

So, to avoid unnecessary losses, what are the key points that we need to know about the construction of the website?
1: domain name registration
First, we need to register a good domain name. Actually, the shorter the domain name is, the closer to the company name, the better. But now there is basically no domain name, so registering a long, well remembered domain name is also acceptable. For example, our company takes caiyiduo.com as the domain name, and recommends that you register.Com domain name. Of course, if your business is bigger and afraid of others, it will be more important to register more.Net,.Cn and.Com.cn. Of course, it's not necessary to spend more money on those more expensive domain names.
2: the server
Now a good server can directly link up with the ranking of the website. The faster the user opens your website, the better the cheaper server may be unstable. For instance, it will open very slowly, and the website will not be able to open. This way, if the time period is very unlucky, spiders are crawling away from search engines, which is very unfavorable for the ranking of websites. So buying servers is better. Now the server is not very expensive, thousands of yuan a year is a very good server. Buying domestic mainframe needs to be put on file, but when buying foreign host, the ranking is not good at home, and the speed of access will also be affected. The general enterprises choose to build domestic hosts and make foreign trade companies.
3: Website Design
The content of the website is inevitable or need, many companies will choose the website template, but also has a small part of the selection from the custom website establishment, template website focus is cheap and station fast, but the emphasis of custom website promotion and brand building. There are competition in the industry. Users who do template websites must consider less than that, and template website optimization is not as good as customizing websites. From programming, the same has some disadvantages, and search engine algorithm is updating, so website code is optimized, template website should take into account many aspects, so many code is useless for enterprise station. In terms of design, because customers choose template website, the Internet company will never help you redesign. At the very least, we will design 3 banner advertisements for you, help you to add some information, and template website is built. And the custom website is well designed from the design, the general user is satisfied, the website can be online. So the design of a custom website must be the advantage.
The key is website later stage if do not carry on the marketing promotion, then the website does not have the significance, if sells the promotion, the website design is also important, therefore looks at the user to consider in which aspect, is choosing the website construction direction.

數據中心網絡帶寬線速有門道

數據中心：線速是體現網絡設備轉發性能的一項重要指標,很多數據中心在採購網絡設備時,都會要求設備具備線速能力,或者是部分線速能力。在RFC 1242中對線速做了定義:在不丟失任何一個幀的情況下最大轉發速率,以太網吞吐量最大理論值稱爲線速,若是千兆設備就要達到千兆的線速,若是萬兆設備就要達到萬兆的線速。理論歸理論,實際中畢竟要考慮很多外界的干擾因素,因此很多時候的網絡都是“僞線速”,線速是要在特定情況下,實驗室環境中才能測試出來的。在選型設備時,不必一味強調滿足各種情況下的線速,實際中除了廣播風暴,也不會設備所有端口都做線速轉發,那一定是一個異常的網絡狀態,一般端口速率超過80%時,數據中心都會開始網絡擴容了,根本不會等到端口線速轉發時纔去擴容。線速是理想化的東西,在選擇設備的時候千萬不要看重這個,否則會被欺騙的。就像我們購買的小汽車,最大行駛速度都可以達到200多公里∕小時,而很多人一輩子都不會將汽車開到那麼高的速度,小汽車設計的理論速度的確可以達到,但會有各種條件限制,所以對網絡設備線速的事兒,且不可鑽牛角尖。下面,就來詳細說一說網絡設備爲滿足線速性能裏的門道。

網絡設備有的只有1 u高,有的卻有20 u高,對外端口有千兆,萬兆,40克、100克甚至更高,尤其是在框式設備上,不同端口速率的板卡插在同一個機框裏,要全部滿足所有板卡都線速,是很難做到的。因爲低速端口板卡內部需要的連接器只要是低速的就可以,而高速端口板卡內存需要高速連接器,在一個機框裏很難全部滿足,或者在某些板卡組合的情況下,部分端口就無法達到線速。這種情況在早期的網絡設備中表現更爲明顯,那時內部連接器速率都比較低,內部還不是信元轉發,按照報文哈希轉發,內部很容易出現擁塞導致業務丟包。在這種情況下,如果數據中心要驗證採購設備的線速性能,往往設備商會將能線速部分展現出來,而小部分無法線速的部分儘量在測試中避免。還有隨着測試報文數量越大(或數據幀越短),網絡設備需要處理和校驗的負擔就會越重,出口轉發速度必然要下降的,但是越接近線性關係。很多網絡設備在大包數據的處理上是可以達到線速的,而報文長度越小,達到線速的難度越大,若報文只有64字節,對設備的性能考驗是最大的。在這種情況下,網絡設備可能達不到線速。其實,我們知道實際的網絡中,是不可能只有一種64字節長度的報文,肯定是各種長度的報文混雜在一起,這時對設備的壓力還不算最大。
線速的概念主要指的是交換機網絡設備,這種設備靠硬件芯片轉發,可以具備線速能力,而這些設備的CPU處理能力是比較弱的,所以CPU處理的報文是遠遠達不到線速的。交換機的CPU不會處理轉發數據報文,除非硬件芯片裏沒有了轉發表項的情況下才會考慮通過CPU轉發,交換機的CPU主要是協議報文的處理和設備管理,處理報文的能力相對很弱,沒有線速的概念,就算是硬件芯片也不是什麼情況下都能滿足所有端口線速,有的芯片受工藝水平所限,芯片整理的轉發帶寬就有瓶頸,當所有外部端口都線速轉發時,芯片就會有丟包,芯片只能保證部分端口線速情況下無丟包。我們知道,在很多選型測試中,經常使用蛇形測試,即將面板所有外部端口都收尾相連,打入線速流量,看是否有丟包,很多設備在這種情況下無法測試通過,就是芯片本身存在端口線速的數量限制。還有路由器,它是靠CPU轉發數據的。路由器雖然CPU能力很強,但是要滿足線速還是很困難,一般路由器會考慮用NP芯片來完成數據轉發,或者也植入硬件芯片來完成,靠硬件的處理速度來滿足線速的轉發,這種設計理念使得路由器和交換機的界限越來越模糊。在很多時候,有人拿路由器當交換機用,有人拿交換機當路由器用,讓兩者技術實現上不斷融合。
線速的測試其實是有標準的,RFC2544就是線速測試的標準.RFC2544明確建議40歲,64128256512年,1024年,1280年和1500年字節這些數據幀是需要測試的。在線速的流量情況下,測試網絡設備的丟包,時延,抖動,吞吐量,背靠背,這些概念在網絡上都可以找到在此不再細說。這裏要注意的是,以太網報文有能看到的部分,也有看不到的部分。在以太網報文之前,還有96位的空閒幀。空閒幀是根據以太網的CSMA / CD原理,用來偵聽鏈路是否空閒,如果空閒,就可以發送報文。接着會有七個字節的前導碼AA(01010101)用於與接收端同步,因爲電平一高一低,很容易同步。最後還有一個字節的AB,作爲幀界定使用,表示後面開始真正的以太幀,這20個字節在日常的網絡抓包中是不可見的,是以太網物理層封裝的東西。這部分長度並不一定是固定的,而是可調的,有的設備爲何提升轉發效率,將這部分調小,這樣單位時間內轉發的報文數量就多,轉發效率提升,不過,這樣做也會帶來困擾,就是與其它設備對接時,其端口轉發速率要比對方高,可能會出現超線速的情況,對其它設備有影響。線速是要求設備提供的幀間隙和前導碼按照標準默認的來,有的網絡設備小包達不到線速時,往往將這部分長度調小,還提升線速能力,這樣就與線速的標準背道而馳。
線速一般是網絡產品在實驗室裏的理論值,是理想化的東西,在實際應用中也要達到線速是比較困難的,所以在選購網絡設備時,不要過於依賴這一性能指標,還是要看網絡設備的綜合處理能力。

Data center migration: data destruction technology in data centers

Data center migration: data center to be concentrated area of information processing, produce huge amounts of data every day, and these data not only takes up a lot of storage space, also affect the application of computational efficiency.
In fact, most of them are useless data in the data, garbage data or intermediate data calculation, data center saved these data and no benefits, there are some of the timeliness of data, outdated data, make the data in the data center storage devices in the sleeping, as clean out, saving storage space, so can greatly save cost of data center to improve production efficiency of data centers.

How to destroy useless data, there is also a lot of knowledge here.
First, data quality is far more important than the amount of data, data center in order to increasing data and the practice of continuous investment in hardware and software, was a costly mistake, should pay attention to the effective data processing.
On the surface, the amount of data data center growth is very rapid, global data volume is growing at 58% a year, less than two years has doubled, the speed will be faster in the future, most of these data are generated in the data center.
Our data center is not possible to double the capacity expansion and every two years, so let the data volume growth, data center will soon fall in continuous expansion of the cycle, the result data center size bigger and bigger, and data center operations are not substantial growth, profit decline in data center.
It's like being a fat man, always eating Fried chicken cokes, getting fatter and fatter, but actually the body's physical fitness is falling, and in the end, you can't do anything but proud flesh.
Data cuts cannot be the burden of data centers, which should be cleaned when cleaned and destroyed when destroyed.
Second, we need to destroy the data, and figure out which data is useful and which data is useless.
This is about to start from the data source, the incremental data in the data center, to classify the data, compile, store into the classification of storage space, for specific tag name of the data at the same time, through the name of the data is known about the content, so that as judgment criterion of the data is useless, if you are not clear when data records, then the later data destruction will not be able to do precise, not only inefficient, destroy data will likely also useful data destruction by mistake.
Management of these data is very complex, involves data identification, cleaning, optimization, and so on, and the work it is cyclical, takes time and a certain human resources, and will not bring significant profits, is often overlooked.
In fact, the effective storage of data will generate long-term and positive benefits to the business of the data center, and the earlier the benefit becomes obvious.
Third, the destruction of the data is not simply the removal of the deletion. The data destruction is also standard to follow.
DOD's DOD 5220.22-m standard is the most widely used set of regulations, with many people using the DOD 5220.22-m as the standard for data clearance and destruction.
There are several methods of destruction, which are related to the effect of destruction.
General destruction is divided into soft destruction and hard destroy two kinds.
Soft destruction is data destruction or data erasure by data coverage and other software methods.
Hard disk data destruction is destroyed by means of physical and chemical methods to destroy the storage medium directly to achieve the purpose of complete hard disk data destruction.
Soft destroyed generally will be destroyed, the data file is not the real disk area will wipe out data, operating system, due to considering the operators operating habits or wrong operation, data after the destruction of all kinds of situation very much, and many other factors, the delete command users use, just put the file directory entry do a delete tags, put them in the file allocation table occupied cluster cluster marked as empty, not to make any change data area is not do any of these information data erasure, data destruction operation, actually still occupy storage space, such data did not achieve the goal of saving storage space.
So data center data destruction is hard to destroy, the storage space is actually released to store more meaningful data.
Hard to destroy common methods: format hard disk, hard disk partition, file shredding software.
Format is only for the operating system to create a new empty file index, all sectors is marked as "not used" state, data disposal data erasure that the operating system don't think the hard disk file, therefore, if the use of data recovery tools software can recover after formatting data in a data area, the formatting is advanced formatting, low-level formatting, quick format and partition formatted several.
The destruction of low-level formatting is the most thorough, and it is difficult to recover the destroyed data through the software, and the storage space of the hard disk is fully released.
Hard disk partition ways of destroying data, only changed the hard disk master boot record and system boot sector, most of the data area and has not been modified, not achieve the goal of data disposal data erasure.
Crushing software is specially used to delete file to achieve data disposal data erasure, there have been many on the net, some anti-virus software also increased the function of data disposal data erasure, private data can be used to handle general, but cannot be used to handle with the classification of the data.
The above method can destroy the data, but it is not safe enough, or it can be recovered by the malicious people and get the data to do wrong.
Still have a kind of hard to destroy is using special degaussing hard drives or bending machine to thorough destruction of the data, either directly to demagnetization of hard disk, or bending of hard disk are destructive behavior, the hard disk will be damaged, not only can't restore data, hard disk cannot be used again, this kind of hard to destroy is often used to handle has failed hard drive, avoid faults in the hard disk to save the data to be bad reduction, holds the data to do bad things.

機房建置氣流遏制如何兼容部署的各種機架

機房建置氣流遏制並不像採用環評標準設計的數據中心機架那樣得到廣泛應用。雖然這其中有一些行業沒有采用的因素,但關於成本高昂和沒有效果的說法是完全站不住腳的。氣流遏制將始終爲數據中心提供良好的環境,可以支持更高的功率密度和更低的能源成本。

但現實是,數據中心的氣流管理面臨着如物理,機械或硬件的複雜性等一些問題和障礙。事實上,除非企業從一開始就積极參與數據中心的,設施,建築工程,以及戰略規劃的新設計,否則將在部署氣流遏制設施時將會面臨一些棘手的問題。這是因爲數據中心的消防系統,管道系統、梯架,電源母線槽,托盤等設施和設備對於數據中心氣流管理來說是一些主要的物理障礙物,而大多數數據中心沒有專門的供應和返回的空氣路徑。
在引入數據中心氣流遏制技術之初,工作人員改造現有數據中心空間經常遇到的問題是由於數據中心機架具有不同的尺寸或形狀,很難適應數據中心的新設計,不符合遏制兼容安裝基礎的外部物理規範。由於這些障礙實際上是新興技術面臨的問題,所以這也是數據中心解決方案提供商和第三方首先要解決的問題,需要解決多家廠商的機櫃產品的多樣性差異的問題,並保證貨源充足。
數據中心通過增加氣流通道的塑料圍擋和圍簾實現早期改造。這些系統可能與專門設計的系統一樣複雜,具有高架地板支持的基礎設施和滅火啓動釋放機制,並封堵從頂部供應管道到機櫃頂部的橋接間隙。在這種情況下,修剪塑料條或窗簾的長度就可以進行調整。儘管這種方法很簡單,但早期的採用者也獲得了一定程度的效率和效果的改善。自此,氣流遏制供應商提供各種各樣的方法和產品以適應不同的方案,並努力填補各種柔性插頭所有的孔洞。
由於今後還將引入更多不兼容的機櫃或機架,這將導致數據中心高效的遏制系統受到破壞,其應用前景令人擔憂。然而,隔離這些機櫃改造成的櫃式煙囪的設計解決了這個問題。因爲大多數所有的密閉供應商都提供分區選項和插孔附件,爲兼容性集成提供了一種途徑。
最後,無論是數據中心面臨改造的問題,還是面臨當前或未來物理障礙的新設計,局部氣流遏制都是實現多個供應商機櫃遏制的一種有效途徑。無論是解決方案,如排門的末端還是機櫃頂部的部分分區,供應商都推出了一些解決這些問題的解決方案。

How does website design choose space for you

For some starter website design, how to choose the procedures to build websites and what kind of space to build websites is a headache. In this way, digital technicians come to answer for you, hoping to help some new webmasters.
First of all, the operating system of the building space should be considered.
The operating system is the most important of the web server, and learning how to choose the suitable operating system is of great help to the establishment of the site. At present, the operation system of server is mainly divided into two systems: Windows and Linux, each has its own advantages and disadvantages. Windows system is convenient for operation and visual interface for new webmaster. Linux system needs to have a certain understanding of it, pure command operation server. But compared to the security and system stability, the Linux system is superior to the Windows system. As a server, security and stability are also very important. If a server is often attacked by a hacker and causes the site to be inaccessible, who will dare to use it?

Secondly, which language programs are supported by the server
The server supports what language, mainly to see the server operating environment, the current mainstream website program development language HTML, PHP, ASP,.NET, JAVA etc., for individual webmaster general use "HTML, PHP, ASP development of the site is more, enterprises may make use of any of the above is a kind of language. HTML for the operating environment is all support for ASP and.NET, it can only use the Windows system to run IIS environment, for PHP and JAVA have no limits, their Apache environment is cross platform deployment, ASP and.NET do not support cross platform.
Then, how to choose space capacity, database capacity, and space traffic restrictions
The choice of site space is very important, have enough, but as time goes on, upload pictures and other data more and more space due to lack of capacity, if only a display of the website, does not need to upload pictures and information, 500M-1G space is enough to use, if it is a picture video, or upload the information site, the best selection of initial slightly larger space, then slowly back to the expansion of space capacity. Database, 300M is large enough, 50M space can be enough to save thousands of articles, but what you need to understand is to reduce the pressure of database. The best articles will generate HTML static pages stored on the server. Now many host spaces have limited traffic to prevent a large amount of traffic from the host and affect the normal use of other users' websites.
Finally, the number of concurrency connections
The number of concurrent connections is the same time a number of user requests to the server, the maximum number of concurrent users is the server at the same time the server can response the number of connections, if long time exceeds the number of server resources may cause congestion, the site open speed slow, or even a server crash. Note that the number of concurrent connection of the site on the number of visits, but at the same time the server's maximum number of responses, such as the maximum number of connections with support for 150, said the same time in support of the 150 open web page, 151st people will need to wait in front of someone after opening the page to visit. For enterprises with low traffic volume, the number of concurrent sites should not be too high. More than 150 of them are enough. For the active community forum mall and other websites, large enough concurrent numbers are needed to support the operation of the website.

数据中心配電系統的选择及控制

1 數據中心配電系統架構及等級分類
典型的數據中心供電系統由中壓配電、變壓器、低壓配電、不間斷電源、末端配電以及發電機組等設備組成，其中UPS的主要作用是在市電電源中斷、發電機啓動之前，確保所帶負載的持續供電。

2  數據中心對發電機組的要求
數據中心持續功率（DCC）在柴油發電機組行業內的應用持續增長。
DCC應用作爲數據處理中心（DPCs）電源的一種可替代電源，它的特性使它能滿足數據中心設施對可靠性和可用性的需求。
Uptime Institute規定應急電源需滿足TierⅠ或TierⅡ，替代電源需滿足Tier Ⅲ或Tier Ⅳ要求。
要被認爲是可替代電源，並滿足Uptime Institute Tier Ⅲ和Tier Ⅳ的要求，則該發電機組必須在主電源發生故障時能提供持續電力。
這意味着發電機組容量能在要求的負荷水平下提供持續電力而不受時間限制。
可以從以下幾個方面來選擇最適合
數據中心項目的發電機組：
（1）主電網的可靠性、數據中心所存數據的敏感度
如果主電網足夠可靠，或數據中心處理的數據敏感性要求較低，或不嚴格要求數據的實時可讀取性，那麼數據中心設施的有效性百分比將降低，它們會被歸類爲Uptime Institute Tier I或Tier II.這時會根據發電機組的備用功率來選擇發電機組作爲備用電源，以減少安裝費用。
（2）發動機及發電機的選擇
作爲電力供應的動力源，發動機的重要性不言而喻；
不但要考慮通用的環境因素（溫度、海拔）、啓動冗餘（雙啓動馬達、雙啓動蓄電池，或氣動、彈簧啓動）、通風散熱、發動機水加熱等；
還要結合數據中心的要求來選擇發動機及發電機。
發電機組行業主要的發動機製造商新發布了一個新的功率定義——數據中心持續功率（DCC）。
發動機生產商用DCC定義來保證發動機運行不受時間和平均功率百分比限制。
發電機組的連續負載能力、可靠性、突加載能力是最核心的要求。
（3）智能冗餘控制系統
控制系統作爲發電機組運行的大腦，肩負着自動開關機、同步並聯帶載、負載分配（有功、無功功率均分）、功率管理、保護、數據傳輸等功能。
所以控制系統要做到充分可靠及冗餘控制（控制冗餘、通訊冗餘）。
3 發電機組冗餘控制及案例分析
冗餘控制器是一種熱備份的應用，防止系統崩潰導致機組不能正常運行，在實際應用中當用主控制器出現故障時備份控制器可以無縫接替主用控制器的當前工作狀態，確保系統高可靠運行。
冗餘控制的重點在輸出信號、輸入信號以及參數設定；
冗餘系統中的主、備用控制器通過CAN總線通訊；
備用控制器通過CAN總線週期性的向主用控制器發送信息，以評估主用控制器的狀態。
當主用控制器出現問題時，備用控制器會通過二進制輸出點來控制外接繼電器，瞬間將主用控制器的輸出信號切換到備用控制器的輸出口上，主用控制器故障到備用控制器投入時間最多200ms.
4 結束語
數據中心處理的數據越敏感，就越需要可靠的電力設施爲設備提供持續電力。
由於發電機組是構成這類設施的一個基礎部分，所以在選擇數據中心發電機組時要依據Uptime Institute制定的標準來選擇，依據備用功率來設計容量大小時，需滿足Tier I和Tier II要求；
滿足Tier III和Tier IV的要求，發電機組則必須依據它的DCC功率來選擇容量。
同時發電機組智能控制系統需具有熱備份、冗餘控制功能，確保供電系統的可靠性、穩定性。

機房建置氣流遏制如何兼容部署的各種機架

機房建置氣流遏制並不像採用EIA標準設計的數據中心機架那樣得到廣泛應用。
雖然這其中有一些行業沒有采用的因素，但關於成本高昂和沒有效果的說法是完全站不住腳的。

氣流遏制將始終爲數據中心提供良好的環境，可以支持更高的功率密度和更低的能源成本。
但現實是，數據中心的氣流管理面臨着如物理、機械或硬件的複雜性等一些問題和障礙。
事實上，除非企業從一開始就積极參與數據中心的IT、設施、建築工程，以及戰略規劃的新設計，否則將在部署氣流遏制設施時將會面臨一些棘手的問題。
這是因爲數據中心的消防系統、管道系統，梯架，電源母線槽，托盤等設施和設備對於數據中心氣流管理來說是一些主要的物理障礙物，而大多數數據中心沒有專門的供應和返回的空氣路徑。
在引入數據中心氣流遏制技術之初，工作人員改造現有數據中心空間經常遇到的問題是由於數據中心機架具有不同的尺寸或形狀，很難適應數據中心的新設計，不符合遏制兼容安裝基礎的外部物理規範。
由於這些障礙實際上是新興技術面臨的問題，所以這也是數據中心解決方案提供商和第三方首先要解決的問題，需要解決多家廠商的機櫃產品的多樣性差異的問題，並保證貨源充足。
數據中心通過增加氣流通道的塑料圍擋和圍簾實現早期改造。
這些系統可能與專門設計的系統一樣複雜，具有高架地板支持的基礎設施和滅火啓動釋放機制，並封堵從頂部供應管道到機櫃頂部的橋接間隙。
在這種情況下，修剪塑料條或窗簾的長度就可以進行調整。
儘管這種方法很簡單，但早期的採用者也獲得了一定程度的效率和效果的改善。
自此，氣流遏制供應商提供各種各樣的方法和產品以適應不同的方案，並努力填補各種柔性插頭所有的孔洞。
由於今後還將引入更多不兼容的機櫃或機架，這將導致數據中心高效的遏制系統受到破壞，其應用前景令人擔憂。
然而，隔離這些機櫃改造成的櫃式煙囪的設計解決了這個問題。
因爲大多數所有的密閉供應商都提供分區選項和插孔附件，爲兼容性集成提供了一種途徑。
最後，無論是數據中心面臨改造的問題，還是面臨當前或未來物理障礙的新設計，局部氣流遏制都是實現多個供應商機櫃遏制的一種有效途徑。
無論是解決方案，如排門的末端還是機櫃頂部的部分分區，供應商都推出了一些解決這些問題的解決方案。

How the data center migration should be implemented

Data center migration is not an easy task, but it is a problem that enterprises must face, because data centers need to merge, transfer, integrate, build and update. There is no doubt that such a task will soon become a new challenge for data center administrators, especially when there is need for ground layouts and other physical needs.

In addition to a very clear plan, such as setting up the training period of the machine, arranging the old equipment to eliminate and the server integration, administrators must also consider the downtime of the application software. Fortunately, the complete cessation of data center services is currently avoidable because of the availability of data center hosting and low hardware prices.
However, the application of software hosting needs to be fully planned, no stage and product requirements can be wrong. Although the core system is converted to the external environment has been greatly reduced, or sometimes administrators have to bite the bullet choice escrow way so as to ensure continuous operation of application software.
Proper planning is the key to the smooth transition of services and devices, and only in this way can the user's work be affected without any impact. Finally, the end of this arduous task of the administrator will require good communication between different IT teams, including engineers and technicians, so that the redeployment task can be successfully completed.
Recently, the CMP path detection center laboratory (CMP Channel Test Center Lab) needs to be migrated. In order to successfully complete the migration, we must create a small plan, which will ensure that only two to three days, the infrastructure can't achieve maximum utilization. The ground design, the supply of power resources, the USP service and the network design have changed when the whole plan is half completed. In order to minimize the downtime during the transfer period, we quickly developed a plan of action. The following are the implementation steps of the plan. The solution providers can formulate similar transfer plans, so that they can avoid the occurrence of problems in the transfer process.
1, coordinate the equipment, electricians and IT staff. From the very beginning, we clearly told every group that we needed to transfer the minimum amount of infrastructure equipment and rack. The new server machine room is smaller than before, so we are faced with the potential space arrangement of the required rack and shelf. We calculated the minimum amount of rack and rack required - both to meet the needs and to be placed in a new machine room.
2, the shutdown but the data center continues to run: on the day of the transfer we put forward an ordinary but very effective action plan. Employees continue to open key network paths when they transfer equipment and components.
3, cooling systems and other systems: there is a neglected content here. Because of design reasons, the CMP Channel Test Center new server room requires less cooling facilities, but the total workload of the data center has not changed. We need to observe and measure carefully. When transferring a data center, you also need to discuss with the electrician the related questions, for example, the maximum power load that the new computer room can get, so as to ensure that the number of future machines can grow, but also make CIO and CEO understand the power bearing capacity of the computer room.
4. Maintenance of data center operation: data migration and guarantee of the operation of core application software are never a big problem. However, according to our experience, the least cost action plan is to decompose the work. In other words, there are two data centers that actually exist during the migration. The VAR that maintains a small data center should be recommended in this way to ensure the continuous operation of the data center. This can be achieved without the need for infrastructure hosting during the migration process.
5, start: not in cannot but under the condition of not breaking the frame and network connection. The less the disconnected connection, the less the cost, and a client can recover more quickly. In order to speed up the migration speed of CMP Channel Test Center, employees try to ensure the connection of the line as far as possible, and attach labels to the disconnected network connections. The result shows that we have saved a lot of time.
6, test, test and retest: do not take care of any small link. Detect network connections, external services and servers, keep in touch with other people involved in migration, and make them responsible for any damage.
The laboratory re awoke us deployment. A lot of small details that are often ignored in a lot of work and device testing show their importance and need to be paid attention to in the migration.

Enterprise website design How to match colors

In the enterprise portal the website design, with respect to collocation and use of color, should be scientific and pertinent. It is necessary for us to know and master some knowledge about color matching. In the face of different cultural background of the enterprise website, the use of color should be different. Choose a suitable and unique color matching to highlight the corporate cultural personality. Design with clever collocation of color, can make the site more exquisite beyond compare.

1 optimize the design of the enterprise website
In the era of Internet information economy, the only way for the development of enterprises is e-commerce. Therefore, it is a crucial issue for enterprises to enter the field of e-commerce. The electronization and networking of business operations are highlighted by the booming development of the leading e-commerce websites such as Alibaba, Jingdong, Dangdang and so on. The trend of trade is the trade of goods through the assistance of Internet technology. Therefore, the design of the website is particularly important, and its image represents the image of the enterprise.
1) users: users can feedback to enterprises through the use of websites, and timely absorbing and receiving reasonable and useful user experience to upgrade and optimize their website can greatly enhance the attractiveness of websites.
2) based on the search engine promotion site: search engines can quickly obtain information to optimize website design and guide users to find relevant information. Click search, enter the website, get information and service until it becomes a real customer.
3) website operation and maintenance: after the website optimization design, the enterprise website can be consistent with the network marketing strategy, and it has the real network marketing orientation. The upgrade of Web site operators for web management and maintenance technology can also play a role in the use of network marketing and the accumulation of more marketing resources.
2 color of the website
However, users browse a website, first impression, not the rich and colorful content and unique layout, color is. The color changes in temperature, weight, hardness, size of visual color can make people have rich psychological association, so, if we can skillfully use and color collocation is a website for the first time can grab your attention, or even leave a key psychological suggestion.
2.1 the color of enterprise website
1) company's exclusive: the Pizza Hut Inc web page hue is a warm tone - red, because the eye - catching red meets the appeals of the fast food store to make the passers-by go into the store. The dark blue of China Mobile, the grey and white of Apple's official website, and their advertisement and posters are also the same color. This also explains that each company has a standard color to represent the company's CI image.
2) company style: the color of the website serves the style of the company. The main color is blue with blue group, not only highlights the core value of the company, to delixin winning, but also promote a sunlight transparent, like-minded, competitive growth and efficient execution, customer satisfaction, charity culture. The mixture of blue and white on the website can reflect the noble, atmospheric and elegant atmosphere. It is the unique style of the enterprise. Iqiyi video website mainly use green, green and white collocation, make people feel comfortable, highlight the company has always enjoyed the quality of the concept style.
3) personality: the color style of the individual website is determined by personal preferences, reflecting the individual personality of the human being.
The classification and significance of 2.2 colors
What is easy to ignore is that color can actually have a great influence on human body and mind. The extreme coloring and coloring will give a strong psychological suggestion and even guide some of the people's behavior. When we are proficient in the different roles of different colors, we can guide the visitors to a certain extent.
Red is one of the three primary colors that can stimulate the excitement of the nerve fibers. It makes people feel like passion, desire, strength and love. The Yellow brightness is the highest, allowing people to enjoy optimism, sunshine, hope and happiness. Orange makes the psychological balance and warmth. Purple is mysterious and noble, but it also feels negative and lonely. Blue is a typical cold color, suitable for high tech modern sense of the IT industry. Green is between the two colors of cold and warm, which is the representative of nature, health, youth, vitality and richness. Brown symbolizes the earth, home and hearth, represents the reliability, comfort, endurance, simplicity and stability. And black is the symbol of wealth and the devil, with the feeling of elegance, the feeling of fear and the mystery.

冷通道模在資料中心的應用優勢

數據中心建設規模日益增大，採用大型的冷通道模解決方案，系統雖然複雜，但是能效高、節能效果明顯，日益成爲主流。

數據中心基於自身的應用特點，對離心式冷水機組有着更高的要求，比如要求機組更加穩定可靠、節能、避免喘振、快速重啓等等。
而變頻離心式冷水機組契合了數據中心的應用要求，目前已經在數據中心行業廣泛使用。
下面我們看一下變頻離心式冷水機組在數據中心中的應用優勢：
1、  變頻機組能效更優
IDC數據中心冷水機組設計多爲N+1冗餘設計，結合變頻機組熱備模式運行，即使數據中心IT設備負載率很高的情況下，冷機絕大部分時間也在部分負荷狀態運行。
以某品牌某款機組爲例（冷凍水供回水溫12/18℃），機組在80%負荷百分比、不同冷卻水進水溫度條件下的情況下，變頻機組與定頻機組能效對比：
可以看出，變頻離心冷水機組在部分負荷下效率比定頻機組高很多。
並且，隨着冷卻水進水溫度降低，變頻機組的節能效果更加明顯。
2、  提高可靠性
2.1啓動電流低
定頻機組爲星三角啓動。
一次啓動電流高達滿負荷電流的200~250%，二次啓動電流甚至可能高達滿負荷電流的500%.
變頻機組爲軟啓動，啓動電流不會超過滿負荷電流，減少了設備的電流衝擊，延長設備壽命。
2.2 快速重啓
對數據中心而言，短時間的供冷中斷不但會造成室內熱量的大量堆積、溫度迅速升高，還將導致IT設備性能降低、使用壽命縮短等一系列更爲嚴重的後果。
所以爲了應對短時斷電或電源切換等極端情況，數據中心通常要求冷水機組在儘可能短的時間內實現重啓及加載。
變頻冷水機組沒有啓動時間間隔的限制，機組可頻繁啓停，機組在帶UPS快速啓動的情況下，可實現在1分鐘之內重啓，2分鐘內到達80%負荷輸出，定頻機組難以實現。
2.3避免喘振
喘振主要是因爲冷凝壓力與蒸發壓力壓差過高或壓縮機制冷劑流量過低。
定頻機組在低負荷運行時，只能選擇調節導流葉片開度，很容易發生喘振現象。
變頻機組在低負荷狀態運行時，同時調節導流葉片開度和電機轉速，調節機組運行狀態，可控制離心機組迅速避開喘振點，避免喘振對機組的傷害，確保機組運行安全。
當然，給機組配置熱氣旁通避免喘振會更加放心。
3、  其他
變頻機組還有其他優點，比如部分負荷運行時噪音降低、提高機組功率因數、減少備用發電機容量、如果配置機載啓動櫃還可以節省佔地面積等。
但同時，變頻機組的變頻器也會對通信及數據中心配電網產生高次諧波。
高次諧波不僅造成設備自身和電網相當大的附加無功電能，而且會干擾通訊系統，嚴重情況會造成系統死機、控制失常停機，甚至會丟失實時數據，影響設備正常運行並且縮短使用壽命。
因此，對於冷水機組機載變頻裝置要求配置濾波器。
尤其是IT設備與變頻器在同一個變壓器下，應配置有源濾波器。
隨着離心式冷水機組在大型超大數據中心中的廣泛使用，變頻技術已經比較成熟並深入人心。
但變頻機組絕大多數爲單級壓縮機組，我們也期待着多級壓縮與變頻技術的有機結合，通過更多的案例看看會有怎樣的效果。
與此同時，考慮到機組的節能降耗、穩定可靠，變頻磁懸浮機組也越來越受關注，也期待其精彩的表現。

政務大機房建寘謹防一哄而上

機房建置是信息時代支撐政府部門日常運轉的重要基礎設施，對提升政府部門信息化水平、推進“互聯網+”政務服務發展、提高政務大數據開發利用能力，以及推進社會治理能力和治理體系現代化都具有重要意義。

這兩年來，我國各地對政務大數據中心建設亦非常重視，都在不斷加強規劃設計、加緊推進建設，但同時，一哄而上，重複建設、投資浪費、產能過剩等現象也十分突出，亟需規範。
時下，地方政府紛紛推進政務大數據中心建設。
從省級層面看，在全國31個省市自治區中，已有12個基本建成省級政務大數據中心，15個正在建設，4個正在規劃。
從服務屬性來看，其中24個設計爲單一的政務大數據中心，7個設計爲“政務+行業+產業”雲平臺。
從投資模式來看，已基本建成的12個省級政務大數據中心中，有6個由政府直接投資，4個由省國有企業出資，2個由省國有企業和其它企業合資建設。
從地市和縣區層面看，全國有40%以上的地市、20%以上縣區，正在或計劃開展本地的政務大數據中心建設。
地方政府在政務大數據中心建設方面能夠投入極大熱情，其原因在於將其作爲地方推動政務應用和促進產業發展的重要基礎與抓手。
一方面，地方政府希望通過建設政務大數據中心爲政務雲、政務大數據等應用提供信息基礎設施，支撐基礎數據庫、主題數據庫和業務數據庫等政務信息資源體系建設，推動政務信息系統統籌集中建設，推進政務信息資源共享交換和公共信息資源開放，進而釋放政務數據紅利，全面提升地方經濟社會的發展水平。
另一方面，它還希望基於政務大數據中心拓展社會服務能力，發展交通、物流、旅遊、工業、商務等各類行業應用，推動行業服務提檔和傳統產業升級，激發大數據領域的創新創業熱情，打造本地雲計算、大數據產業生態圈。
然而，由於有些地方政府對於政務大數據中心建設前期缺乏實際、深入和詳實的調研工作，並且對大數據及雲計算應用的特點、規律瞭解不夠，使得政務大數據中心建設初步顯現出應用需求未明，盲目確定建設規模政務大數據中心的服務器數量規模已成爲各地政府追逐和比拼的重要指標；
重建設輕運維，建設成效難以發揮；
企業介入過多，規劃設計不盡科學等三方面問題。
爲此，應加強對政務大數據中心建設項目的立項審批管理。
不但要編制政務大數據中心投資建設指南，爲各地建設政務大數據中心提供建設規模測算、造價評估、體系架構、安全保障等方面的參考與經驗借鑑，還要建立政務大數據中心項目立項審查工作機制，由政府辦公廳、發改委、工業和信息化等部門聯合加強項目管理，嚴格審查項目建設需求和規模，杜絕同類項目的重複投資建設，確保基礎設施類和基礎資源類項目統建共享。
同時加強對項目可行性研究論證，按照“隨機、公開”的原則，隨機抽取論證專家評估的政務大數據中心項目投資模式、建設模式、運營模式、運維模式等，並將論證結果上網公開，長期接受社會監督和意見反饋。
應完善政務大數據中心建設項目的評估、考覈和審計機制。
除了建立政務大數據中心建設項目全生命週期評估機制，加強對項目建設需求、可行性研究、投資規模、實施方案、服務能力、運維質量、經濟效益和社會效益等的全方位評估，還要建立政務大數據中心建設項目責任機制，明確項目建設、承建、運營和運維等各方責任，強化項目責任人的首要責任和終身責任機制。
此外要建立政務大數據中心建設項目績效考覈機制以及政務大數據中心建設項目審計、倒查和問責機制。
應推動政務大數據中心項目建設和運營模式創新。
比如鼓勵系統集成商牽頭，聯合基礎硬件平臺建設商、網絡服務接入提供商、軟件信息服務運營商等產業鏈關鍵環節企業共同參與建設，構建利益共享和風險分擔機制，明確各方的任務分工和責任，構建共同運維服務保障機制，避免發生一旦出現技術問題就相互推諉的情況。

Data center migration, location is the key

Data center migration sites play an important role in our decision making process. From choosing where to live, where to work, where to spend a holiday, and even where to do business and how we manage our information.

The growth of data, together with the need for more flexible and customized IT capabilities, means that a lot of money is needed to manage and maintain the internal IT infrastructure, which is becoming harder to prove. On the contrary, third party online stores, such as hosting facilities, and the flexibility of the terms provided by these providers, now means that you can transfer responsibilities to others, so that you can rest assured that your IT infrastructure is secure. But once you decide to outsource, how would you choose a provider?
Historically, London is usually the best place to store information, but with the consideration of rents, scalability, and data growth, operators are starting to think more, so there are more locations to choose from. Recently, a report from BroadGroup, a data consultancy, showed that Ireland was the best place to establish facilities in Europe, and the reasons for it were the connection between cities, tax incentives and government active support. Among them, Amazon and Microsoft have offices in Dublin, it is worth mentioning that, here is one of Microsoft's largest offices in europe.
Now, apple is expected to set up a 850 million euro data center in Athenry, outside Dublin, which makes more and more companies start to think about the traditional data center position. So what are the factors that need to be taken into account when looking for a location?
First, facilities in major cities automatically involve costs and risks associated with urban life. In rare places like London, the rental market is very competitive, so the cost is passed on to the customer.
Another consideration is the future cost - whether your business is likely to grow, which means you may need more rack space. If this assumption is established, it needs to ensure that the center can provide room for growth, because it is expensive to relocate the facilities to the new environment. The centers outside London tend to be bigger, have greater development and scale, so it is easy to integrate into new technologies.
Although location related costs may be specific to a business, security and risk are a common concern in the industry. Traditionally, data centers are often close and connected to each other. In London, you'll find most of the data centers are in the east of the city. Although the proximity to the financial district and the major exchanges has obvious advantages, but at the same time, there are also potential risks, especially to put all data centers in one basket.
People are living in a real society where terrorism threatens, and the recent attacks in London and Manchester are a serious reminder, and it also intermittently shows that we can't predict the coming of an event. The city that attacks may be locked in, and remote guidance measures can reduce the loss of this problem, but what happens if your data center has a problem and no one will solve it?
In addition, the natural environment of the data center is also an important element to define security, which essentially increases or reduces the risk of the enterprise. The data center is located near the river, such as Thames River, and in the case of flooding, Zhejiang directly affects the operation of the data center. Locating a data center in a flooded area has always been a dangerous strategy, and the thing that needs to be protected is that once the flood defense system fails, the entire IT system will be destroyed.
For big cities, fires are also a constant threat. Although today's advanced fire protection systems have a long history of protecting urban data centers, the risks are still very high compared to those outside the data centers located outside the city.  When looking for data center partners, these are things that companies should consider. For each company, each business has its own unique needs, its facilities to meet these needs is very important. Even big cities like London seem to be the choice of opinion when outsourcing information, but it's important that it's not the only option. This has been confirmed by the rise of Ireland and other regions. Knowing this will help you gain more features and benefits, reduce the risk of risk, and significantly reduce costs.

Website design must ensure the safety of the website

I think a lot of programmers in the website design will set up a multi-channel anti hacker attack program, because the web design is not good, a very important indicator is that the website is not safe, will often be hacked.

As the basis of network marketing promotion, the website has been more and more attention by many enterprises. Not only the design is becoming more and more beautiful, but also from the user's point of view, so that the site's sensory experience is getting better and better. In fact, this should be so, for the vast number of enterprises, the purpose of website is very simple and very direct, in order to obtain more users' attention, so that the brand image and marketing promotion. Compared with the website design, interactive production and marketing flow data which can be directly perceived, the website security problem has been hidden in the website operation and promotion, and it is not easy to be found and not received due attention. As everyone knows, website construction and marketing, lose security, lose everything. Compared with the past, in recent years, search engine traffic hijacking and web content encountered tampering has a high incidence trend, this article tries to explain the importance of website security from three aspects of website data storage, content transmission and regular security maintenance.
Ensuring data storage security
What is the concept of data storage? Popular speaking, the web page files, databases and pictures on our website are placed on a server, this store. And to ensure the safety of website data storage, is to protect the website files and databases, etc., remove the normal content updates, not malicious tampering, malicious use. In most cases, the occurrence of Web site data leakage or web pages encountered tampering, is the site of data storage has led to hidden trouble. One is the site server operating system vulnerabilities, open ports too much or security configuration problems, hackers led to. Another reason is that web site programming is not standardized, using data SQL, such as or other ways to lead to web tampering. The security of website data storage depends more on the professionalism of website construction personnel and technical maintenance personnel. Generally speaking, it is more reliable to entrust professional website construction company.
Ensuring the security of content transfer
The essence of the Internet is the electronic network which is connected with each other. When we send a request in a certain place, a lot of routing nodes can intercept our request content by a certain method as long as there is a certain intention and a certain technology. This is easy to test, we installed a packet capture software Wireshark in local computer open operation began to capture, and then open a non encrypted website, input the user name and password core data, you can find our own username and password can be grasped. And that's why we've seen a lot of financial websites, enabling HTTPS to transmit web data through SSL encryption. The author thinks, if there is a member system or trading system on the page, and the technical conditions permit, it is best to carry out SSL encryption finishing on the whole station.  Compared to thousands of dollars a year, digital certificate fees, to ensure the safety of the site is more important.
Regular, effective and safe maintenance
Some companies believe that as long as the site is ready to buy a server casually, and then upload it can be opened. In fact, this kind of idea makes many companies website lack of necessary security maintenance, most of the time in the state of streaking. The server operating system used by the website is developed by people, and the code of the website is knocked by the programmer's line and line. Where is the absolute permanent security? Once the vulnerability is discovered and the vulnerability is announced, the website is in a very dangerous situation. We have seen a lot of web pages have been tampered with, and most of them are old websites for many years, the fundamental reason is that these sites lack the most necessary security maintenance. A website development specialist believes that regular security maintenance and data backup are essential. If the company has professional IT personnel, it is best to carry out a careful examination of the website every month and vulnerabilities, virus scanning killing. If the lack of technical team, you can entrust professional website construction company to maintain, but do not let the site neglect maintenance, in the streaking state.

光纖系統如何規範跳線管理

對於光纖系統來說，電信間及設備間是數據、語音、圖像三類業務的匯聚地，其重要性不言而喻。

但是對於它們的整體設計、設備定型、硬件配置、施工維護等各方面下足了功夫。
但是，施工方往往會忽略了電信間及設備間裏面數量最多的設備維護和安裝保障-電纜、光纖跳線，而忽略這個問題，會給我們的機房管理工作帶來很多麻煩。
因此本文認爲有必要針對跳線進行正確地管理操作。
一般來說，合理的跳線管理可分爲5個階段：計劃、準備、配線、測試、驗證。
跳線操作規範
1、計劃
預則立，不預則廢，做任何事都需要事先做好詳盡地計劃。
針對跳線管理來說，應做好現在和未來的需求規劃。
（1.1）變更請求。
各種管理活動、移動、添加或更改（MAC）均始於變更請求。
變更請求必須含有啓動規劃程序的所有必要信息。
（1.2）搜索記錄。
收到請求表後，應對記錄進行搜索，以確定所用電路路徑。
（1.3）正確路由。
確定正確的跳線長度前，首先要找出待連端口之間的最佳路由。
通常爲通過水平和垂直纜線導管的最短路由，而且不得阻礙或妨礙配線架中的其他跳線或連接器。
選擇跳線、應避免過度鬆弛，確保外觀整潔。
跳線太緊會增大對連接器的拉力，而過度鬆弛則會給跳線管理帶來麻煩，增加配線架的管理難度。
2、準備
做好跳線管理的計劃後，那就應按照事先做好的計劃，接着就應做好跳線管理的準備工作。
在實施管理操作之前儘量多做準備，研究管理記錄。
確定需要連接和重新連接端口的位置及相關端口的標籤信息。
（2.1）先檢查需要跳線的型號，然後再檢查該跳線的質量情況。
爲了確保跳線質量正確無誤，就需檢查跳線是否損壞，爲了檢查其是否損壞，當然先可從跳線外觀來查看，如果有條件的話，可用專業的儀器檢查。
（2.2）接着檢查需要連接部位的情況，以此來避免連接部位的物理損壞。
（2.3）最後需要對跳線接頭和連接部位的清潔。
對光纖連接器的清潔有接觸式和非接觸式兩種方法：
接觸式清潔方法：
（1）擦拭紙及無水酒精，採用原生木漿配以特殊加工工藝，超低粉塵，質地純淨，高效吸水，紙張細膩，不會刮花被插拭物表面，用低塵擦拭紙配合無水酒精對光纖連接器進行擦拭；
（2）無紡布，不產生纖維屑，強韌，不帶有任何化學雜質，絲般柔軟，不會引起過敏反應，而且不易起毛和掉毛，是作爲光纖連接器或插針生產或測試時清潔用擦拭布的理想選擇，在使用過程中要配合無水酒精對光纖連接器進行擦拭。
（3）清潔棉籤，專門設計用於陶瓷套管內部清潔，或者用於清潔法蘭盤（或適配器）內不易到達的插芯端面；
（4）專業清潔器，光纖連接器專用清潔器採用專用成卷的擦拭帶，裝在可捲動的外殼中，無需酒精，每次清潔都非常有效併產生一個全新表面，方便實用。
非接觸式方法：
（1）超聲波清洗法，它將清潔液變成超聲“液柱”送到連接器端面，並在同樣小的空間內將廢液回收並吸乾淨；
（2）高壓吹氣法，它的原理是在連接器端面先塗上清潔液，然後用高壓氣對準連接器端面吹；
（2.4）檢查光纖連接器清潔情況
清潔完光纖連接器後，都必須對端接面進行檢查。
一般做法是使用100、200倍或400倍的放大鏡進行檢查，下圖是顯示出光纖端接面在清淨的狀態和被污染後的狀態。
跳線管理人員，不管用何種方法，對於一些嚴重污染的連接器還是很難清洗乾淨的，需要配合使用棉花棒及酒精等清洗液加以處理。
經過這一系列的準備工作，那意味着跳線管理的配線工作即可展開。
3、配線
配線架的安裝，應根據操作規程完成各個階段的任工作。
跳線施工中紐結、毛刺、箍縮和接觸不良均有可能大幅降低跳線性能。
若要避免此類問題，應重點考慮以下因素：
（1）彎曲半徑
跳線允許的最小彎曲半徑需要遵守跳線廠商的操作規範。
標準規定，非屏蔽雙絞線（UTP）的最小彎曲半徑應爲纜線直徑的四倍，屏蔽雙絞線則爲纜線直徑的八倍。
2芯或4芯水平光纜的最小彎曲半徑爲大於25mm，如果彎曲半徑小於此標準，則可能導致導線的相對位置發生變動，從而導致傳輸性能降低。
（2）跳線拉伸及應力
配線過程中，請勿用力過度，否則可能加大對跳線和連接器的應力，從而導致性能降低。
（3）捆紮
跳線不一定都需要捆紮，如果捆紮需要遵守廠商的捆紮原則，不要捆紮過緊，否則會引起對絞線變型。
請勿過分擰緊線夾，應以各條跳線能自由轉動爲宜。
請使用專用產品，考慮選擇無需工具即可反覆使用的產品，如尼龍粘扣帶。
4、測試
（1）雖然經過了跳線配線完成，但可能並非認爲該光纖鏈路或銅纜鏈路是否完全符合操作規範或者綜合佈線國際和國家標準，那就應該進行光纖或銅纜測試，只有在符合測試標準後，纔可斷定是否通過測試標準。
5、驗證
（1）花些時間對連接進行最後可視化檢查是值得的。
確保跳線鬆弛處未紐結、未被機櫃門夾住。
（2）最後一步是根據現用配置更新記錄，關閉與已經執行完畢的變更請求相關的工單。
現在跳線已經是綜合佈線系統中重要的組成部分之一，尤其是數據中心項目中跳線的良好管理操作，就顯得尤爲突出。
相信只要通過施工管理人員正確合理地跳線管理操作，必將使得整個綜合佈線成爲真正意義上先進性、科學性、實用性、可靠性的系統。

機房建置運維的水平發展路標

運維是機房建置的重要工作，機房建置一旦建成，後期要經歷一段漫長的運維期，期間不僅要保證業務的平穩運行，還要不斷對系統進行升級和擴容，以便數據中心可以不斷開展新的業務。

所以，數據中心對運維的工作都異常重視，運維的水平高低反映出了這個數據中心整體業務水平的高低。
隨着數據中心領域的蓬勃發展，對運維的工作提出了更高要求，運維的工作也需要持續改進，去適應新形勢，數據中心發展的需要。
本文就來詳細講一講數據中心運維的水平發展路標，看看高水平運維的工作體現在哪些方面。
數據中心運維的發展原則有兩個方面：一個是儘量不去依賴人去管理，要知道數據中心裏百分之八十的故障是人爲故障，人蔘與程度越高的工作出錯概率越高，反而機器永遠都按照預定的程序去執行，除非設備出了BUG，否則永遠都不會出錯，當然BUG也是人造的，所以往往一個數據中心自動化運維的水平越高，反而越安全，故障發生的概率更低；
另一個是要儘量避免發生故障，而不是事後諸葛，“亡羊補牢，爲之晚矣”，不要總去做亡羊補牢的事情，要把可能預知的風險消除掉，避免故障的發生。
故障發生後，迅速解決故障是一種能力，但不要過於依賴這個，不能什麼問題都要等到故障發生後纔去解決，早早就應該規避風險。
“覆水難收”，故障發生後給數據中心帶來的負面影響，往往要花更多的精力去修復，有時發生的故障是致命的，數據中心可能會從此一蹶不振，只能關門大吉了。
任何一個數據中心運維的工作，都要依照這兩個原則去發展，這樣才能不斷提升其數據中心運維的水平。
數據中心運維的水平高低也可以從兩個方面來看，一方面是運維效率，另一個方面是規範建立機制。
首先，在運維的效率方面，從低到高要經歷四個階段：一是全人工運維。
這種運維的方式適用於早期數據中心規模不大或者業務流量不大的情況，這類數據中心繫統複雜度不高，設備數量較少。
日常的業務運維操作，更多的是依靠手工逐臺登錄設備進行操作，缺少必要的操作標準、流程機制。
運維的人員個人經驗非常重要，可繼承性不強，數據中心要過度依賴個別的幾個運維的技術大牛來維持，其它人員操作犯錯概率會增高，同時工作效率底下；
二是工具化的運維。
這種運維的方式適用於較大規模的數據中心，運維的人員開始使用批量化的操作工具，針對不同操作類型出現了不同的腳本程序，需要做設備配置變更時，通過腳本程序統一執行，提升操作效率。
比如設備批量升級，可以提前寫好腳本程序，然後到了指定時間，腳本程序自動運行，將服務器上的軟件程序下載到設備上，然後執行升級命令，所有設備的執行步驟都一樣，可以大大節省人力，以往人工升級每晚只能升級幾臺設備，通過腳本一個晚上就可以將整個數據中心的設備升級完畢。
不過，每次操作需求都不同，需要不斷調整腳本工具，可程序化處理能力較弱，批量執行還可能導致更大規模的問題出現，此時仍需要人工監督腳本執行情況，發現腳本有問題及時調整，運維效率並不高；
三是平臺運維。
這種運維對運維效率和誤操作率有了更高要求，通過平臺承載標準、流程，進而解放人力和提高質量。
平臺運維對服務的變更動作進行了抽象，形成了操作方法、服務目錄環境、服務運行方式等統一的標準，通過平臺來約束操作流程；
四是自運維繫統。
這種運維適用於更大規模的服務數量、更復雜的數據中心繫統，是當前數據中心推崇的運維方式，極大地解放人力。
自運維繫統對服務變更進行抽象，由調度系統根據資源使用情況，將服務調度、部署到合適的服務器上，自動化完成與周邊各個運維繫統聯動，比如監控系統、日誌系統、備份系統等。
自運維的系統還具備發現故障，並自動消除故障的能力。
另一方面是規範機制的建立。
俗話說“沒有規矩，不成方圓”，數據中心裏也要立規矩，制定各種規章制定，並有效地執行下去，規範的建立同樣也要經歷從低到高的四個階段：一是無規範機制，整個數據中心運維的工作處於無序狀態，工作效率低下，這在一些小型的數據中心或機房普遍存在，過多規範制度反而顯得有些累贅；
二是建立規範人工約束，這個階段通過規範制度加強對人的管理，通過規範人的操作流程，從而減少人爲出錯的概率。
數據中心制定了一系列操作規範，哪些不能做，哪些可以做，哪些人能做哪些事兒等等，運維的人員要按照規範來執行；
三是完善規範，不斷對規範進行改進，防止出現管理漏洞，運維的工作開展遵從一系列規範制度，有理有據去做，提升數據中心的運維效率，對運維的人員獎罰分明，依據就是這些之前制定好的規範制度；
四是系統自動約束，此時數據中心已經完全採用自運維的系統方式，人工參與極少，所以以往制定的一系列規範制度成爲了廢紙，我們只要將標準的操作輸入給運維的系統即可，系統可以自我調整，自動運行完成，保證不會出現不符合規範操作的情況。
數據中心建設的規模越來越大，採用人工方式已不現實。
要將所有運維的工作都能走向自動化，減少人的重複工作，使我們的運維交付更高效、更安全。
數據中心運維的技術發展宗旨就是將人從複雜枯燥的運維工作中解脫出來。
數據中心的所有運維活動，均由人工處理變成系統自動實現。

Data center mobility policy: what needs to be considered in the transition to SDDC?

When the data centre migration, it takes a lot of time and careful study to move your traditional equipment onto the large platform. In mobile data center, what strategies should be kept in mind?

SDDC: a virtualization and software technology for all physical and hardware resources in the data center. SDDC relies on virtualization and cloud computing technology, SDDC is the goal of all the physical resources data center virtualization, virtualization technology, construction consists of a virtual resource pool of resources, not only is the server virtualization also includes storage virtualization and network virtualization etc.. Not only can simplify the server changes, storage changes, network configuration difficult, but also makes the server, storage, network management and configuration operations with repeatability and sustainability.
SDDC enables hardware resources to be configured and scheduled by software, improving flexibility and agility, and a significant advantage is that it greatly reduces the cost of data centers. With the centralized software management layer provided by SDDC, management becomes simpler. At the same time make SDDC software to manage the network, make the network become a part of the data center, a proprietary network can greatly improve the efficiency of the hardware, thus the importance of software can not be ignored. Software defined data centers will become a new direction and trend of data center evolution.
So, when you've decided to start building or to SDDC (software defined data centers) over, you need to think about some issues. For example, when you're on the hot channel of the data center at the moment, you might be frustrated. How do you deal with the existing devices in the data center and the applications that run on them? The only answer is: like most things in information technology - "it depends on the situation."".
For mobile data centers, two basic models should be considered: first, run some old and new ones in parallel, or integrate your existing devices with the new SDDC in some form of transformation. Second, the integration of existing network devices into an independent data center, wrong, not concentrated to one, is two, or integrated in the pod layer, or integrated in existing devices running SDDC top layer.
The first model is to run the old and new data centers in parallel, which may seem simpler, even desirable. But even in the ideal case, there are some problems to be solved. Will workloads be transformed between data centers? If so, how does this happen? Of course, the answer to this question largely depends on the application itself. There are several important questions to be asked in this field.
How does the new data center architecture meet the requirements of each specific application? It is important to consider the issues that are usually considered, such as bandwidth utilization, delay and jitter (Jitter is composed of deterministic content and Gauss (random) content. The so-called jitter is a kind of jitter. How do you explain that? Its definition delays from the source address will be sent to the target address, there will be a different delay, such a delay change is jitter.) and other needs. But considering the domain name system, the dynamic management of elephant traffic, the creation of security domain, overlay network and other factors are also very important.
Elephant flows: in computer networks, elephant traffic is a very large (total byte) continuous traffic that is established by measuring TCP (or other protocol) traffic on the network link. Although the flow of elephants is not very large, it can occupy a disproportionate share of the total bandwidth over a period of time. It is not clear who created the "elephant", but on the Internet the word published in 2001 in the beginning to happen, when observed, a small amount of traffic carrying most of the Internet traffic, the remaining traffic flow contains a large number of these network traffic is very small (mouse flow). For example, researcher Mori et al studied the traffic volumes of several Nihon University and research networks. In the WIDE network, they found that the elephant traffic accounted for only 4.7% of all traffic, but during this time occupied the 41.3%. of all data transmission
The actual impact of elephant traffic on Internet traffic is still a field of research and debate. Some studies have shown that elephant flow is likely to be highly correlated with traffic peaks and other elephant flows. There are different definitions of elephant flow among researchers, including over 1% of the total flow in a period of time, measuring the duration of the flow, and observing the magnitude of the flow greater than the mean, plus the standard deviation of traffic during this period. One of the main goals of elephant flow research is to develop more efficient bandwidth management tools and Internet prediction models. For example, researchers focus on providing better service quality for small size (small flow) traffic by prioritizing traffic flows.
Even in the initial design stage, it is best to avoid some technical problems related to the late service provided by the data center. In fact, in the process, there are always some omissions.
Because there are very few application developers who know what services the application is relying on, or make invalid assumptions during the execution of such lists, at least they don't have complete inventory.

Website design and browser compatibility problems and Solutions

For website design developers, in order to give users better online experience, to solve the browser compatibility problem is still a big challenge. And web design is a combination of technology and art, in the designer to consider the beautiful, but also consider notebook, tablet PCs and mobile phone compatibility issues, web design has shown new trends.

1 browser compatible with web page problems
As is known to all, the Internet is usually through the browser to achieve, the so-called browser is to display the web server or file system HTML file content, to ensure that users interact with the file. The kernel between different browsers is different, which leads to the same page in different browsers in the effect of differences, and even can not be displayed properly. At present, some website design has not been able to take into account the ability of various browsers, through a small number of browsers open web pages will appear deformation, can not access, display incomplete and pictures unchanged. For this problem, web developers to design a good site to be placed on different browsers to detect their compatibility, the emergence of different situations should be addressed by targeted methods.
Most web designers use CSS to deploy layouts. At present, CSS3 divides the CSS into different modules, the function is also continuously powerful, the homepage design is also more convenient, regardless of is the mainstream portal website or each kind of small company even the individual small station, also carries on the design through the CSS. Once, IE occupied the mainstream browser, but with the continuous development of Internet technology, various browser was blowout trend, such as Baidu, Sogou, 360 speed in its own browser, but also occupied a large market share, at the same time, Google, Firefox 3435, other browsers in the market occupies an important position. The kernel used by different browsers is different, which causes many web browsers not compatible, because the browser kernel is responsible for interpreting and rendering web pages syntax. Therefore, the browser's kernel is different, the interpretation of the web page syntax is also different, the same page in different browsers display is also different, this is what we call website design and browser compatibility issues. If the compatibility problem of web page and browser is not good, it may cause the browser to interpret the content of the web page error, appear garbled, deformation, information disorder and other phenomena, affect the beauty and use of the page.
2 solutions to the compatibility problems of web pages and browsers
2.1 using Hack technology to achieve browser compatibility issues
The so-called Hack technology is the use of CSS style different browsers support different features for different browsers repeat the definition of many different styles, from their analytical support to implement their own browser style, to design a different browser with the same display page. The most commonly used method is to use browsers to support the selection of special characters or individual styles, and not to define different styles repeatedly. A special display style of individual browser, if individual browsers have their own separate support hidden style, first for the most general browser definition style, and individual style after the hidden browser support separate duplicate definition of the style, the most browsers use the former browser specific hidden style cover after using the latter alone. If the individual browser does not support the most browsers use style, first for individual browser definition style, then the individual browser does not support the style for most browsers repeat the definition of the style, the individual browsers use the former, most browsers cover after using the latter.
2.2 inconsistent margins of different browsers
For example, in CSS write a margin - left:588px, after testing, people now IE8 and Firefox browser display the same effect, but IE6 display will be a problem, the main performance is that the margins will be a few pixels, which affects the beauty of the page. The reason for this phenomenon is that different kernel interpretations of web pages lead to different rendering mechanisms. Different vendors have different interpretations of CSS, and different versions of the same vendor may also have different interpretations. As mentioned above, the rendering of the same problem by IE7 and IE8 is different. In addition, browsers and CSS and versions have been dynamically updated, which is a factor that often causes the two to not be compatible.  To solve this problem, you can write different standards for different browsers.
When you do web design, designers are usually designed to implement different levels of menu settings. In some delicate web sites, if the mouse arrow is pointing to a navigation site, the arrow will display the hover effect. This kind of display is not a problem in IE7 and IE8, but it can't be compatible when it is opened by IE6. If IE6 is to achieve this effect, you must use JavaScript to write functions to assist the completion. This requires the creation of a hover.htc file that uses the JS script to define the style of the element. If hover is detected, the onmouseout and OnMouseOver events are set to the element to achieve the effect of hover. Since then, the use of hover in IE6 will not be a problem.

數據中心機房網絡佈線施工要點

網絡佈線具體的方式有“田”字形和“井”字形兩種,其中“田”字形較適用於環形機房布的局,“井”字形較適用於縱橫式機房佈局。它的位置可安排在地板下和吊頂兩個地方,各有特點,下面分別討論一下佈線過程中需要注意的問題:

1,構架設計合理,保證合適的線纜彎曲半徑上。下左右繞過其他線槽時,轉彎坡度要平緩,重點注意兩端線纜下垂受力後是否還能在不壓損線纜的前提下蓋上蓋板。
2,放線過程中主要是注意對拉力的控制,對於帶卷軸包裝的線纜,建議兩頭至少各安排一名工人,把卷軸套在自制的拉線杆上,放線端的工人先從卷軸箱內預拉出一部分線纜,供合作者在管線另一端抽取,預拉出的線不能過多,避免多根線在場地上纏結環繞。
3,拉線工序結束後,兩端留出的冗餘線纜要整理和保護好,盤線時要順着原來的旋轉方向,線圈直徑不要太小,有可能的話用廢線頭固定在橋架、吊頂上或紙箱內,做好標註,提醒其他人員勿動勿踩。
4、在整理,綁紮,安置線纜時,冗餘線纜不要太長,不要讓線纜疊加受力,線圈順勢盤整,固定扎繩不要勒得過緊。
5、在整個施工期間,工藝流程及時通報,各工種負責人做好溝通,發現問題馬上通知甲方,在其他後續工種開始前及時完成本工種任務。
6,如果安裝的是非屏蔽雙絞線,對接地要求不高,可在與機櫃相連的主線槽處接地。
7,線槽的規格是這樣來確定的:線槽的橫截面積留40%的富餘量以備擴充,超五類雙絞線的橫截面積爲0.3平方釐米。線槽安裝時,應注意與強電線槽的隔離。佈線系統應避免與強電線路在無屏蔽,距離小於20釐米情況下平行走3米以上。如果無法避免,該段線槽需採取屏蔽隔離措施。進入傢俱的電纜管線由最近的吊頂線槽沿隔牆下到地面,並從地面鏜槽埋管到傢俱隔斷下。
管槽過渡,接口不應該有毛刺,線槽過渡要平滑。
線管超過兩個彎頭必須留分線盒。
牆裝底盒安裝應該距地面30釐米以上,並與其他底盒保持等高,平行。
線管採用鍍鋅薄壁鋼管或PVC。

機房建置現代化的7個步驟

機房建置的現代化是至關重要的。機房必須保持技術進步,以免面臨技術落後而不再保持競爭力的風險。如果沒有對安全性的更新,數據中心將更容易遭到更新更復雜的網絡攻擊。近年來,網絡攻擊事件日益頻繁,毫無疑問,那些努力滲透和攻擊數據中心基礎設施的黑客就是非常複雜的惡意軟件的創造者和實施者。

除了安全問題之外,用戶對數據中心的需求也在不斷變化,這迫使數據中心管理人員致力於事半功倍,這意味着他們必須發揮其基礎設施最大的潛力,加快應用程序和信息技術(IT)的發展,同時,其安全運營目標是達到5個9的可靠性,而不能產生任何額外的風險。
數據中心正在發展和演進中,根據最近的一項調查顯示,近80%的企業首席信息官(CIO)在傳統基礎設施的技術和應用方面遇到阻礙。
因爲以往的數據中心基礎設施並不是爲了支持當前的即時消費需求而建立的,並且面臨諸如數據呈指數增長,大數據,物聯網(物聯網),移動,能源成本上漲,以及本地部署,託管,雲計算,邊緣計算等方面組合的新挑戰。日益過時老化的基礎設施根本無法滿足數據中心的戰略需求。隨着現代化技術的發展,創新的敏捷性使數據中心設施能夠更快地響應需求。而隨時掌握新的技術創新有助於提高企業收入,削減成本,甚至爲了提高效率和可靠性而對基礎設施進行調整和改造。
數據中心的現代化使得人們的期望更容易實現,因爲管理人員可以採用更多的新技術。實際上,數據中心自動化將讓組織減少庫存,錯誤,以及額外的人工勞動,同時實現了高投資回報(ROI),並節省了成本。新添加的設備操作更簡單,運行成本更低,能源消耗更少,故障率更低,採用最新的它技術和產品可以節省成本,降低風險,併爲客戶提供更好和更加個性化的體驗。
如果企業考慮實現現代化可能會花費大量的成本,那麼他們也應該考慮如果採用過時的傳統它設備可能要付出高得多的成本。
要想有效實現數據中心現代化,需要開展一些前期的工作,並投入大量的時間和費用,但春長期的回報是值得的。企業可以考慮實施以下七個步驟來實現數據中心基礎設施的現代化和更新,而現代化數據中心可以在彈性環境中保護客戶數據安全。
(1)保持技術更新
淘汰和更換過時設備的能力取決於在整個生命週期內對數據中心每項的資產跟蹤。這從企業接收設備時開始(甚至是在購買之前)直到設備淘汰的那一天。這樣,管理人員就可以瞭解每一件舊設備的確切使用時間,是否在保修週期內,以及何時達到使用壽命等各種信息。
系統地瞭解什麼時候淘汰設備是一個有益的選擇,因爲老舊過時的設備通常需要更多的維護和操作費用。另外,老舊設備生產效率較低,功耗較高,容易出現故障。在數據中主資產的生命週期中,會出現維護成本比替換成本更高的轉折點。
數據中心基礎設施管理(DCIM)軟件可以幫助數據中心人員瞭解即將淘汰的數據中心組件和正在使用的特定設備的能源耗級別,然後在發生故障之前發出警報,以跟蹤這資產。
(2)工作流的過程管理
DCIM工作流可以跟蹤每個數據中心資產的所有工作:執行的工作,執行工作的人員,執行時間等。通過爲這些信息設置一箇中央存儲庫,可以更輕鬆地安排資源,生成工作訂單,並自動確保團隊保持一致的技能。工作流通過增加流程的一致性和問責性來幫助團隊更高效地工作,從而提高生產力。
(3)通過電源故障模擬強化電源環節
根據六研究公司的數據,由於停電和其他干擾,美國數據中心每年的經濟損失高5500年達億美元。因此需要企業確保數據中心的電源強化,並進行災難測試。這個過程允許管理人員瞭解從數據中心到應用程序的整個流程,並且瞭解哪些電源系統爲哪些數據中心資產和應用程序提供電力支持,從而讓團隊知道停電期間可能面臨的風險。
此外,它還允許管理人員查看誰有權訪問電源系統,以及密碼上次更改的時間和位置。
DCIM軟件應具備執行虛擬電源故障模擬的能力,因此數據中心團隊可以確定在特定系統或設備出現故障時關鍵基礎設施會發生什麼情況。它還應提供最後一次實際控制的故障測試何時執行,以及電源故障恢復是否是業務連續性的一部分的信息。
(4)與ITSM流程管理緊密集成
將DCIM解決方案連接到信息技術服務管理(ITSM)解決方案是數據中心現代化計劃中至關重要的一步.ITSM系統(如BMC,HPE和ServiceNow)應與數據中心繫統集成並共享信息。來自ITSM變更管理系統的變更必須傳遞給數據中心執行,這提供了設施和它人員之間的重要聯繫.ITSM和DCIM系統之間的預建連接大大簡化了集成。
(5)採用包含本地部署數據中心,託管數據中心和公共雲的混合策略
數據中心未來戰略是一種混合策略。這種方法的優勢在於其靈活性和適應性。通過實施混合策略,應用程序和工作負載可以在最有意義的地方運行。
當構建最理想的計算環境時,混合策略提供了一個很好的選擇,提供混合的本地數據中心,私有云以及第三方公共雲提供商服務,以及平臺之間的互操作性。混合雲還允許企業隨着計算和預算的變化而提供一定的靈活性,從而使工作負載在私有云和公共雲之間波動。這也使企業可以選擇在需要時部署更多的數據。
(6)工作量分配
靈活性是組織對於工作量的最終目標。選擇運行工作負載的最佳位置實際上取決於組織對這些工作負載的優先級。一個組織需要解決以下這些問題:
是否關心成本效益?
性能對於這個特定的工作負載是否重要嗎?
這些數據是否屬於“邊緣”設施,以使計算和分析更接近最終用戶,從而縮短延遲並提供更好的客戶體驗嗎?
爲了安全,合規或其他原因,管理人員是否更加關注內部部署數據中心工作量?
混合策略可以靈活地移動工作量來實現目標。
(7)對虛擬化/容器層物理集成以降低工作負載風險
虛擬化無疑可以節省資源,並且是一種更爲明智的使用方法。虛擬化允許組織創建計算機網絡資源,硬件平臺,存儲設備等虛擬版本。通過服務器的虛擬化,採用軟件模仿硬件,如CPU的內存或網絡流量或其他設備顯。然,虛擬化系統中的軟件性能水平與硬件水平並不匹配,但是由於用戶不需要使用所有的硬件,因此具有更大的靈活性和控制能力。
但是採用虛擬化,用戶通常不知道設備物理層的位置,也不知道物理地址。通過將DCIM集成到虛擬化層中,用戶可以全面瞭解每個虛擬服務器上運行的內容以及工作負載和應用程序在哪個物理硬件上運行。通過這種可見性,管理員可以查看最重要的工作負載的運行情況,從而保護這些特定的物理服務器。
數據中心實現現代化並不是一個難題,而是什麼時候實現的問題。數據中心的設施基礎設施必須保持最新狀態,否則將會落後。現代化的數據中心環境將會更安全,更經濟,更有效,並且更加適應新的數據挑戰,提供商業機會,能夠爲內部部署的數據中心和雲端客戶提供所需的客戶體驗。
DCIM解決方案仍然是數據中心現代化工作的核心。藉助DCIM,數據中心設施管理人員可以在整個生命週期內輕鬆跟蹤數據中心資產,提高效率,保障安全,提供工作流程,使所有內部團隊可以更有效地協同工作,最有效地分配工作量,並集成其他系統的虛擬層。
而如果沒有部署DCIM,那麼這樣的數據中心就不能被視爲是一個能夠滿足不斷變化的技術生態系統業務需求的真正的現代化設施。