2017年9月29日 星期五

Automatic Transfer system Integrated wiring engineering

The integrated wiring of automatic transfer system is a complex project, and the quality of the project directly affects the performance of the network link. In the implementation of the project, the following points need attention.
 Automatic Transfer system
1, the overall wiring system must be designed in advance, with civil, fire, air conditioning, lighting and other installation works together, so as not to produce unnecessary construction conflict.
2, under the conditions permit, the weak should go their own weak wells, reduce the chance of electromagnetic interference, floor wiring room and mainframe room should be arranged as much as possible for the development and maintenance of the required. For networks, the laying of physical layers is critical because it is the foundation.
3, try to 'at some point, the double panel (a speech, a data), with power with the good, in the information near the source point of cloth. Because it is a one-time in the project of integrated wiring line in general, good cloth, to change the wiring is difficult, but with the development of communication equipment is more and more, so some of the more secure duobu.
4, don't make integrated wiring too idealistic, don't put some of the more specific network also integrated into the integrated wiring. Integrated wiring, the telephone network, computer networks, building automation network integration is enough.
5, along with the rapid development of network equipment, the selection scheme should be open, by more than five or five categories, to ensure the data and voice can be exchanged by optical fiber cable vertical logarithmic increase, for some large buildings, should be used to do data backup instead of using optical fiber cable.
6, because the light product prices continue to drop, large bay wiring should be considered 'points of light so favorable for network construction.
7, laying DG pipe to use larger diameter, to leave margin. Cable should pay special attention to the turning radius, turning radius is too small will cause serious loss of check each link, a fiber optic cable, light box panel special contacts, some of the panel of the box is not enough depth to spot after doing not installed on the panel to the box is good, installed after the test is not good, the reason is installed after the cable radius is too small, causing serious loss.
8, with the price of light welding equipment, to local conditions, can give ST and SC method of making traditional head into the fiber pigtail and phase welding practices can ensure optical quality.
9, not one-sided pursuit of wiring products brand, in recent years, in fact, the performance of network equipment non shielding wire, some domestic manufacturers of cable, module on other has reached the industry standard, the price has a larger advantage, so we can consider the use of their products.
The above points are frequently encountered in engineering. In fact, the integrated wiring is not a difficult to understand the project, as long as the principle of understanding it, a lot of comparison of product performance, we will be able to make affordable systems.

現代機房建置服務器維護檢查清單

機房建置服務器只是複雜的機器。
像任何機器一樣,它們需要定期維護,以達到最佳性能。
簡單的維護程序可減少嚴重的服務電話,延長服務器的使用壽命。
即使現代服務器擁有強大性能與豐富功能,增加的工作量整合和可靠性預期可能會對您的業務造成損失。
 機房建置
您的服務器維護覈對清單應包括物理元素以及系統的關鍵配置。
堅持常規檢查
服務器管理員也經常忽略規劃維護窗口。
不要等到發生實際的問題,再預留時間進行例行的服務器預防性維護。
維護頻率取決於設備的年齡、數據中心環境、需要維護的服務器的數量等因素。
例如,位於設備倉庫中的較舊設備需要比採用HEPA過濾方式部署的、冷卻良好的數據中心中的新服務器更頻繁的檢查。
組織可以根據供應商或第三方提供商例程的日常維護計劃,如果供應商的服務合同每四到六個月要求進行系統檢查,遵循該日程安排。
準備工作決定一切
在處理服務器維護清單上的項目之前,請先制定計劃。
這包括檢查系統日誌中是否包含需要更多直接關注的任何錯誤或事件。
例如,如果系統日誌表示特定內存模塊的錯誤,您應該訂購一個替換的DIMM並將其安裝。
同樣,如果有固件,操作系統或代理補丁或更新可用,請在維護窗口之前先測試並檢查這些補丁。
還需一份明確的將系統脫機並將其返回服務狀態的明確計劃。
在虛擬化技術出現之前,服務器及其駐留應用程序將需要停機以適應維護窗口——通常迫使IT人員在夜間或週末執行維護。
虛擬化的服務器支持工作負載遷移,以取代停機時間,因此您可以將應用程序遷移到其他服務器,並且在底層主機系統上發生服務器維護時,將它們保持爲可用狀態。
在服務之前,瞭解虛擬機應該出現的位置,將虛擬機遷移到所選系統,並確保每個工作負載正在運行,然後再將服務器關閉進行維護。
此時,通常可以關閉服務器並將其從機架或其他機箱中取出。
確保服務器正常運行
一旦服務器脫機時,目視檢查其外部和內部氣流路徑。
除去灰塵和其它碎屑能夠妨礙冷卻空氣的任何累積。
從外部空氣入口和出口開始檢查,隨後進入系統機箱,查看CPU散熱器和風扇組件,內存模塊和所有冷卻風扇葉片和風道通道。
用清潔、乾燥的壓縮空氣清除適當、靜電安全的工作區上的灰塵或碎屑。
不要在機架上打掃服務器。
打掃灰塵是一個古老的流程,但這並不意味着它已經過時了。
灰塵是一種絕熱材料,如何去掉它們的方法非常重要,現在替代的冷卻方案和ASHRAE建議已經提高了數據中心的運行溫度。
灰塵和其他氣流障礙物將導致服務器使用更多的能量,甚至會引起本可避免的早期組件故障。
檢查本地磁盤
許多服務器依靠內部硬盤進行引導、工作負載啓動和存儲,用戶數據和其他功能。
磁盤介質問題嚴重損害了工作負載的性能和穩定性,往往導致磁盤過早故障。
磁介質不完美。
普遍的問題包括扇區損壞和碎片。
RAID在存儲錯誤後保持數據完整性有很大的進步,但較小的1U機架式服務器不能提供足夠的物理空間來部署磁盤陣列。
使用CHKDSK(檢查磁盤)實用程序等工具來驗證磁盤的完整性,並嘗試恢復任何損壞的扇區。
Windows Server 2012的更新版本的CHKDSK可以快速分析和修復文件系統結構中的磁盤問題。
只要NTFS和文件分配表或FAT文件系統首先使用磁盤空間可用的集羣,磁盤碎片根本不會消失。
碎片可能會拖慢服務器的磁盤並導致故障。
Windows Server 2012下的Optimize-Volume等實用程序會將每個文件的集羣連續排列在磁盤上。
閱讀事件日誌的詳細報告
服務器在事件日誌中記錄大量信息,特別是有關問題的詳細信息。
沒有仔細檢查系統,惡意軟件和其他事件日誌,沒有服務器維護清單完成。
當然,關鍵的系統問題應該立即引發IT管理員和技術人員的關注,但是, 無數的小問題可能預示着長期和嚴重的問題。
當您查閱日誌時,請檢查報告設置並驗證警報和警報收件人是否正常。
例如,如果服務器組有技術人員離職,則需要更新服務器的報告系統。
仔細檢查聯繫方式;
如果錯誤發生在工作時間之外,則報告發給技術人員公司電子郵件地址的嚴重錯誤可能完全不夠用(譯者注:非工作時間,技術人員可能無法登陸公司郵箱獲取重要的問題報告)。
主動跟日誌數據。
當日志檢查發現長期或重複的問題時,積極的調查可以在問題升級之前進行問題解決。
例如,如果服務器的日誌報告內存模塊中的可恢復性錯誤,它將不會觸發關鍵警報。
但是重複的實例會導致模塊出現問題,IT人員可以執行更詳細的診斷來識別即將發生的故障。
如果問題不是那麼嚴重,無需關閉服務器,則可以在更換硬件進入之前重新恢復生產狀態。

Optimal planning scheme for data center migration

Over the past few years, I have been concerned about data center migration, and these observations have laid a foundation for my thoughts about the challenges of data center migration today. Now, we have the necessity of future data center migration detailed planning, rather than to avoid the power off sexual assault in the middle of the night for migration, or worse, in the last moment of migration was completely restored. But if you make a good plan, regardless of whether you actually is to avoid or solve these problems, in the course of implementation, relevant experience will play a positive role, but does not need to debug, consider. My executive and sponsorship team is always keen to emphasize, "if this happens, we have the relevant steps and plan to take the appropriate means to solve the problem."."
 Data centre migration
The contingency plans for emergencies and the data center migration are as follows: some suggestions that can help you make some detailed planning ahead of time:
The adoption of new technologies often means other upgrades
If the migration activities involved in the process to migrate to the new hardware, it is usually to update to a newer and better hardware equipment (may bring cost-effective), whether it is specific to the server, storage device, or network equipment. It is a good thing to actively handle scrap equipment and potentially obsolete equipment.
But at the same time, the new hardware is bound to need to support some other aspects of compatibility, upgrades and support. The most obvious is the server's operating system, which usually requires large or small patches, or even firmware updates. But, in fact, this is only the tip of the iceberg. You also need to examine its potential impact on almost all application stacks. These include the database version, the middleware version, and the application itself. These types of upgrades may bring a great deal of burden to your application management team, both from authentication and testing, and may even have a significant impact on your time and project plan.
And then there's nobody looking forward to the future. I remembered those painful experiences in the absence of sufficient migration, pressure test cases, we found that, after migration, for a particular application and hardware combination, a new multi thread processor chip is actually greatly reduced performance of single threaded applications. It's weird. So, we are now asking whether single line or multi threading is necessary to perform appropriate and detailed performance tests before planning.
Do you want to physically move lift-and-shift as the preferred method of migration?
Many of my clients tend to prefer large-scale physical migration as the preferred method. Of course, logically, you don't need to redeploy cables, plugs, racks, and so on, but is that really possible? EH?...... I'm afraid not. For some projects, there are a range of major risks that you need to consider carefully:
Does the supplier need to re test the equipment, especially for larger equipment, such as tape libraries or storage arrays?
Does this involve the choice of shipping companies?
What are you going to do with the data if these data are located in the storage area network (SAN)?
Are you planning to move the entire storage area network (SAN)?
How about the NAS connection?
When will physical transportation be conducted?
Does the server support the security of the internal special data?
If the server is lost or delayed during migration, is there an emergency plan? This plan is definitely not cheap...... You need to insure it.
What changes will occur when the device is reinstalled? Including server name, IP address, firewall rules, backup infrastructure, network routing, DNS, load balancing and so on.
Therefore, based on the above series of reasons, the lift-and-shift migration method is usually the last choice in my best migration selection. Although it may be very effective for special situations such as stand-alone, non production laboratories and test equipment, I do have a lot of bad experience with this approach.

What are the steps in the website design?

Website design is to show the company's corporate image of the specific expression, according to the company's industry, customer requirements, the company's LOGO and other matters needing attention, professional design of a creative creation. Web design is generally divided into the following 6 steps to design:
 Website design
1: some data were collected first according to the customer's industry sectors, such as some of the website to view competition in the industry; then according to the requirements of customers design, in the collection before we need to outline the website sketches.
2: in determining the style and structure of the sketch, we need to collect web information, some of the company's data needs to be provided by customers, such as: website logo, company information, product pictures and so on. Search the web for some of the basic elements of basic web design.
3: according to the design process of web pages is generally at the top, first production site navigation, and then in the design of advertising is what we often say banner pictures, this is a web page display picture is very important. According to the designer's initial sketch, then design the central content section: enterprise introduction, product display, news, information, partners and so on. Finally, we need to design styles at the bottom of the page so that a basic web page is sketched.
4: the basic framework of the site is completed, we need to make minor adjustments, because when you make web pages, we need to establish accurate index, to make a standard site layout. It's about the cleanliness of the web, and when the front-end engineer writes the code when it's making web pages, it's necessary to sort out the pages according to the size and the pixels of the designer.
5: when the home is completed, customers make customer satisfaction after reading, the inside pages, customers feel that the page requires minor adjustments, and then the page designer adjusted according to customer comments, until the customer satisfied.
6: home made sure, designers will start making pages, according to the overall style of the site, making the inside pages, because the page more general time will be longer than the first, probably in 2-3 working days. The best website pages are all confirmed, web page production is basically completed.

2017年9月28日 星期四

Energy efficient data center power distribution system

The power distribution system provides the ability to choose between two operation modes for data center operators: double conversion mode can provide high quality power protection, and multimode or "ecomode" (economic model) with high working efficiency. When the user selects the multi mode as the default operating mode in order to achieve the efficiency of 98% to 99%, if the UPS input is detected on the power supply is abnormal, the UPS system will be less than two milliseconds to automatically transfer to senior protection mode. In the United States, power anomalies that are critical to IT critical loads usually account for only 3% of the business operating time, so the probability of switching between annual operating patterns is likely to be minimal.
Power distribution system
Dual transformation mode is a typical power protection model in large-scale UPS systems for decades, but it also has a significant weakness, that is, the operating efficiency is always in the range of 93% to 95%. Today, users require higher UPS operating efficiency and lower overall cost of ownership (TCO), and can not affect sufficient power protection. This is the key to multi-mode UPS.
A research on the specification of Frost & Sullivan show, a 50000 square foot space enterprise data center using a multi-mode UPS system in 10 years, and most of the time in operation in a multimode mode, can save the cost of $3 million 100 thousand. These costs come mainly from the reduced energy consumption of the UPS system and the energy consumption of the ventilation and air conditioning (HVAC) systems used to overcome the cooling of the UPS system. As the cost of electricity continues to rise, the power savings of multimode UPS will grow as time goes on.
Other multimode advantages other than TCO
In addition to the TCO, the power distribution system offers other advantages and reduces energy consumption. The technology provides a "smart" input power monitoring master, in which the UPS will record the number and frequency of abnormal input power resulting from multimode to dual mode. Excessive or frequent power failure will cause the UPS to "lock" multi-mode for a period of time, so that the power will remain stable for some time. As a result, multi-mode UPS systems can make intelligent real-time logical decisions about when to use multi-mode or dual conversion modes to maximize critical load protection without sacrificing operating efficiency.
The second advantage of a distribution system is that it extends part life. During multi-mode operation, many UPS module components operate at low current and low thermal range, thus prolonging service life. Other components, such as fans, can not run. Manufacturers of multimode UPS expect the maximum life of many components to be extended for one to two years.
The third advantage of a power distribution system is the ease of output failure. Any UPS output short circuit or overload will soon be handled by upstream overcurrent protection devices, and there is no internal mode of operation in the UPS system. These UPS output faults will be directly loaded on the source impedance of the power supply, rather than loaded on the UPS inverter power source, thereby reducing the loss of the fault.
Some UPS manufacturers, such as the Ge Corp's critical power business, have designed additional unique functions in their multi-mode systems, which Ge Corp calls eBoost technology. These features include the economic operation mode of power supply path using line reactor, power adjustment level in the high efficiency mode; specifically, it can reduce the high frequency "surge" events and low frequency "wave", thereby minimizing the conversion for double conversion mode. The design of the reactor also allows the difference of cable lengths between multiple UPS modules to parallel output buses up to + 25%. The cable length difference also provides enhanced floor layout design flexibility. Some other UPS manufacturers must keep the cable length difference at + 10%, otherwise it may cause UPS power down or UPS running in parallel.
The reduction in operating cost (OPEX) of the distribution system and the above additional advantages provide many compelling reasons for the adoption of this UPS technology for data center users in the foreseeable future. For other tasks, critical and / or critical process applications, multi-mode UPS technology should be considered, especially for high-power applications over 225kVA, because energy consumption is a critical issue.

機房建置到底要用多少光模塊?

機房建置高速光模塊市場自2016 年底開始高速增長,其核心驅動力是海外大型/超大型數據中心對40G 以及100G 模塊的升級及新建需求。
機房建置
光模塊作爲實現光電轉換的核心器件,在數據中心內部有廣泛應用。
在10G以上速率、10 米以上傳輸距離的場景下,採用光模塊(或將多模光纖與VCSEL光模塊封裝在一起的主動光纜AOC)是目前唯一成熟的佈線方案。
高清視頻、直播、VR 等新應用推動全球網絡流量高速增長,雲計算、IaaS 服務、大數據等新興應用需求對數據中心內部數據傳輸提出更高要求。
因此,數據中心不僅需要持續擴容、新建,還需要在網絡架構上持續演進,以實現無阻塞的網絡性能。
在數據中心架構演進的過程中,對光模塊的需求產生新的變化,我們對傳統三層架構、改進的三層架構以及新興兩層架構分別進行計算分析可以看出,隨着網絡架構的不斷演進,單機櫃需要配置的光模塊數量顯著增加。
1)傳統三層架構下光模塊數量約爲機櫃數的8.8 倍(8 個40G 光模塊,0.8個100G 光模塊);
2)改進的三層架構下光模塊數量約爲機櫃數的9.2 倍(8 個40G 光模塊,1.2 個100G 光模塊);
3)新興的兩層架構下光模塊數量約爲機櫃數的44 或48 倍(其中80-90%是10G 光模塊,配置8 個40G 模塊或4 個100G 模塊);
從高端光模塊的需求看,海外市場由於雲計算等下游應用的高速發展,第三方數據中心和雲計算廠商自建數據中心快速發展,網絡架構向扁平化持續演進,對高速光模塊產生旺盛需求。
相比之下,國內市場由於下游應用仍處於發展早期,網絡流量壓力以及數據中心內部數據傳輸需求較低,國內數通高速光模塊市場仍需等待下游應用的成熟。
數據中心扁平化主要爲解決數據中心內部(東西向)數據流動,東西向數據主要由雲計算、雲遷移、IaaS 虛擬配置、分佈式計算以及大數據(數據量較大,存儲與不同服務器)等應用驅動。
海外市場AWS、谷歌等雲平臺、Salesforce 等專業平臺以及分佈式計算和大數據應用的高速發展推動數據中心扁平化,對高速光模塊產生旺盛需求。
相比之下,國內下游應用仍處於市場培育和商業模式探索階段,國內高速數通光模塊市場要在下游應用進入成熟期後纔有望進入爆發增長期。

How to control the expansion of data center migration

Data center migration
As more and more technology, resources, and user loads will be put into the data center infrastructure, the future data center capacity will continue to grow. Future demand for data centers will require higher levels of integration, while energy consumption and effective control remain the top priority.
 Data centre migration
Global data center traffic is firmly in the Giga byte era, by 2015 will be more than 1/3, by 2020 will reach 15 trillion and 300 billion bit. Not only the data center traffic continues to grow, but also SDN and NFV and other architectures continue to introduce new, so that their framework is constantly streamlined, for data center optimization, improve the new level. The rapid growth of data center traffic is part of cloud traffic, and by 2020 it will grow to four times now, accounting for 92%. of all data center traffic
Rapid expansion of cloud computing, an important traffic enabler, is increasing data center virtualization, virtualization can provide flexible, rapid deployment and efficient service. By 2020, 92% of the workload will be processed in the cloud.
The interesting part of the evolution around the data center is all components that need to be organically grown through data center extensions. One of the key factors revolves around environmental control. One of the best mechanisms for creating strong data center extension policies is by controlling energy consumption and the data center environment.
With this in mind, let's take a look at how environmental control can help data centers avoid expansion in three ways.
1, create fully customized climate control. In the work of processing data center technologies, many organizations are creating modular, even fully customized data center solutions. With this in mind, a cooling platform that can be modular or even custom designed is critical. Whether it's accurate airflow control or cold air handling procedures - new custom climate control solutions are changing the way data centers expand dynamically. In fact, through the expansion process of the data center, the entire climate control process can be customized to accommodate the needs of very specific data centers.
2, enable application specific cooling solutions. What if you just need to expand a part of the data center? Or - you run only part of the data center, which is a very specific function. The expansion of data centers is driving application specific environmental controls rather than abandoning existing systems. In many cases, you can create containment solutions for the critical part of the data center. Many organizations are actively using powerful containment structures for their critical data center components. The great thing about this is that, by extension, you can tailor these solutions uniquely to a variety of data center application case requirements.
3, environmental control - external, internal and external. Run advanced data center platform. Some of them may be outdoors, some of which may be indoors. The key is that you must have a powerful environment control platform that can be extended with your data center - wherever it is. Outdoor environment solutions can waste energy recovery from waste gas streams, and use it to pre conditioning outdoor airflow, to significantly reduce the heat required to maintain proper temperature and humidity levels in the occupied space, cooling and humidification load. These modular mechanisms allow you to extend data centers indoors and outside. The grommet in the data center, and even DCIM integrated frame airflow management is good method to support the data center environment control.

Website design English website needs to pay attention to a few big points

The website design is now the best way for the development of enterprises, to promote the role of the brand, now in the development of good Internet environment, network marketing is more conducive to the development of enterprises, with the development of foreign trade Baidu optimization bidding, many companies began to build a English website, if you want to open up new areas in the international the market, using the website the corporate brand to the world, also need to pay attention to the following issues.
 Website design
1, website design of Europe and America, simple and in line with overseas customers, reading and usage habits. Note that the size of the font (Westerners love a smaller font), browser encoding (many overseas users not installed Chinese language support website should use UTF-8 encoding), details of the deal (language details, website design fine).
2, the English website design reference foreign well-known big company, make the atmosphere, sedate imposing manner. Foreign users rarely come to the mainland to study, a lot of information through the web site to understand the company, website construction of the atmosphere, can better show their company's strength and professional; company, should show their qualification and strength.
3. placed in overseas trade website server, to ensure that the overseas users'accessing speed, if the globalization of user groups, recommend the use of the host, but must find a very mature overseas server suppliers.
4, the use of overseas mail, or overseas forwarding function of high-end enterprise mailbox, to ensure that global forwarding is normal, the server for spam problem, IP address is often foreign shielding, lead to messages sent abroad, often be returned or not, it will ring the normal foreign trade business serious shadow.
5. foreign trade website construction process, should be the most basic optimization, such as keywords optimization, navigation optimization, tag optimization, keyword density planning must pay attention to these, it can be regarded as the basic configuration, the modern construction site.
6., not to opportunistic, do some dishonest marketing tools, such as sending spam, search engine cheating, the consequences are very serious.
7., learn more about the usage habits of target customer groups, such as in Europe, where many professional buyers use yellow pages more frequently than the number of networks. In some places, the use of B2B business platform more than search engines, and some places are on the contrary, there are exhibitions, etc., according to the habits of different regions to develop different network marketing programs.

2017年9月27日 星期三

What is the Raised floor of the computer room?

Raised floor,In the process of continuous development and renewal of electrical safety technology, it is noticed that a great deal of electrical accidents are caused by excessive potential difference. In order to prevent all kinds of electrical accidents caused by excessive potential difference, since twentieth Century 60 years, the application of equipotential bonding technology has been promoted internationally, and equipotential bonding has been used in new buildings. Thus, the appearance of the potential of the hexahedron of the computer room successfully solved the above problems.
 Raised floor
Hexahedral equipotential earthing of machine room:
The upper, lower, left, right, front and rear faces of the machine room are all metal structures (forming a whole shielding space), and all of them are connected reliably and grounded through one or more points.
Computer room equipotential grounding box; equipotential connection, international attaches great importance to action potential connection, the electrical safety (safety), and lightning protection of electronic information equipment normal operation and safe use, it is necessary to. According to theoretical analysis, the smaller the range of equipotential bonding, the safer the electrical. Concrete construction: the metal component used in the decoration of the computer room is made of plastic copper wire and connected to the equipotential earthing box.
Implementing measures:
1, laying equipotential equipotential pressure belt, in the machine room laying grid type uniform pressure belt, the grid size is about 3M * 3M, material with 40 * 4mm galvanized steel plate, with Phi 8 insulators as support;
2, laying the corner position of bus in each room near the post, set about the two busbar, specifications for 80 * 8mm (four pieces of galvanized steel galvanized steel plate welding), 20 cm long, busbar welding and column reinforcement, the average pressure belt and equipotential connection bus. The power supply PE wire, the equipment shell, frame and other conductive metal objects in the engine room are connected with the bus or equipotential equipotential belt at the same time. The connecting line adopts 25mm2 multi strand copper core wire and galvanized steel plate.
3, laying the grounding grid, laying in the ground deep 1.5M, cross-sectional area of about 4M2 of the grounding grid, grounding grid and the machine room, such as potential voltage with galvanized steel plate connection.
Grounding line laying requirements:
1) grounding should be away from the room where the main building is about 3~5m; 2) digging depth of about 0.8M, 2M long, 2M wide trench in the ground, as shown above, in the location shown uniform placement of 9 bars of 1.4M 2 "(galvanized pipe into a trench about 600mm), and then at about from the ground at 800mm and 300mm respectively welded 12 40*4 galvanized steel; 3) leads to a 40*4 welding of galvanized steel plates in the galvanized plate, the ground is about 1M as a ground connection, the test point; 4) in the grid when welding, the welding area should be more than 6 times the contact and solder joints to prevent corrosion rust treatment; 5) soil by laying resistance reducing agent method (salt, and water) to increase the conductivity and the grounding resistance less than 2 ohm; 6) must be a new backfill clay conductive state better;

機房建置期間要考慮的幾大問題

當你需要機房建置和設計一個新的數據中心時,很容易忽視大局,並陷入細枝末節的問題中。最終導致其設計和建設的結果往往會很不理想,造成工期延誤而需要返工,更糟的是,這將成爲一個長期讓人困擾的問題。
 機房建置
以下是確保數據中心的設計堅持正確方向的五個關鍵問題:
1 .你需要多少個機架?它們的尺寸是多少嗎?
機架的數量決定數據中心的空間需求。傳統的它機架的尺寸爲600 x1000mm(寬x深),而一個空間爲100平方米的機房可以容納大約50個機架。當然機架也有其他的尺寸,而知道機架的尺寸和機架的數量,就很容易估測你的數據中心所需的空間。
2 .你需要什麼等級的數據中心?
數據中心的層等級確定了數據中心的冗餘要求和電源分配路徑。對於一個Tier2級數據中心來說,部署“N + 1”架構的電源和冷卻設備的冗餘是足夠的。然而,對於一個Tier3級的數據中心設施,通常都會包括“N + 1”冷卻設備冗餘,具有雙供電2 N電源冗餘,以及獨立的雙路配電機架,因此,瞭解數據中心的等級可以決定你需要的冗餘設計。
3 .每個機架平均功率密度是多少嗎?
要注意的是,當你設計數據中心的電源容量時,應該根據平均功率密度,而不是在最大額定電源,這是很重要的。機架的平均功率密度乘以機架的數量纔是數據中心的最大它負載量。然後,再加上數據中心的電源系統,冷卻系統和交換機和存儲設備等設施的功率。爲了避免產生昂貴的成本,並充分利用電力系統,使數據中心具有大小合適的電源和空間,採用模塊化策略是明智的,可以降低成本,而不必犧牲未來擴展的靈活性。
4 .數據中心的交流機架或直流機架需要什麼級別的備份電池?
服務器機架可能需要部署百分之百的直流電源,或百分之百的交流電源,或者是二者的組合,例如,如果您正在爲託管目的建設數據中心,你可能有配備交流(UPS)供電系統的需求,而如果是要成爲一個電信設施,這可能需要部署直流電源系統。知道這些,就會知道你所需要的直流電源或UPS電源系統的大小和規模。
在部署備份電池時,最好根據15分鐘備用放電時間進行配置。這樣部署備用電池,不會顯著增加資本支出,雖然這將很難自圓其說,但這更符合成本效益。企業需要提高備用發電機組的冗餘度,而不是在增加電池數量浪費更多的資金。
5 .數據中心將建在哪裏?
分析新建的數據中心所在地的氣候條件是十分重要。這將有助於確定最合適的冷卻措施,其中包括水冷機組,機械製冷空調,自然空氣冷卻,間接空氣冷卻,絕熱冷卻等,這些都是爲了數據中心設施實現其更低的PUE目標。
當然,在新的數據中心的設計中還有許多其他需要考慮的問題。但行業的經驗告訴我們,在數據中心建設和設計過程中掌握這五個關鍵基本設置重點,將確保最終的數據中心設計儘可能接近用戶的真正需要。

Enterprise data center migration principle

There are a variety of reasons why companies need data centers migration. Such as regulatory requirements, mergers and acquisitions, space constraints, or reduced overheads. It is a common event that the business moves from one data center to another, or from one data center to the cloud. However, how to successfully migrate data centers is very challenging.
 Data centre migration
The process involves complex issues, such as redeploy applications, infrastructure software, and sometimes even hardware, and all that they depend on. In the migration process, there are two basic requirements: do not lose any data, nor unexpected downtime.
To achieve these requirements is easier said than done. For example, some applications can communicate through specific connections that have problems when they are far from the new location. In some cases, security and regulatory issues prevent application migration. In other respects, the information on existing IT assets may not be up to date, so it is impossible to predict the impact of some of the hardware or infrastructure elements that have been decommissioned.
The benefits of successful migration of data centers are very obvious. Enterprises can gain greater benefits from new improvements = performance, more efficient management, and migration system operations. Here are six rules to help and guide the successful migration of data centers.
Evaluate existing settings
Corporate knowledge of IT assets is often incomplete. That knowledge is even further down as its assets become obsolete. As time goes on, business process reports, training manuals and internal documents may be lost, while other documents are rarely maintained.
In addition, to ensure that migration is carried out as smoothly as possible, enterprises need to thoroughly check existing data centers. If these information gaps are not handled properly, they will begin to migrate and problems will arise.
Keep in mind that each application is likely to have a variety of relationships and dependencies with other applications and infrastructure elements. The network trace tool helps the staff understand the relationships and re - discover which components are communicating with other components.
However, in the months before the enterprise plans to migrate data centers, the use of these tools makes no sense, as some systems communicate only once a few months or even once a year. So start as soon as possible, and build a comprehensive panorama of data center work before migration.
Select the best migration method for each application
Each application has several different migration methods to choose from, depending on what technology the application is running on. The "promotion and transfer" method refers to physically moving hardware from one location to another, which is suitable for applications that use older technologies.
Other options for applications running on newer technologies include restoring the machine image, copying the virtual machine to the new data center, reinstalling the application, and migrating data. Once the enterprise chooses the most relevant migration method, you need to figure out how many tests the process requires.
It is worthwhile to perform tests and rigorous tests on applications that are particularly important or critical tasks.
Test migration for higher priority applications
When you move a critical task application, it is meaningful to try intensive testing after migration. If an application is less important, such as tracking an employee's birthday application, there is no need to have a trial and test. The decision ultimately depends on how important the system is to the business and what role the technology plays.
High priority applications usually contain sensitive data. To avoid downtime or data loss during migration, companies need to schedule regular backups and develop a solid pre migration rollback plan just in case. Try the migration test method and identify any issues before the migration of the damage activity.
Consider old traditional systems
The IT system may be obsolete, but that doesn't mean it's redundant. Many IT systems still provide good service to enterprises after decades of installation. The challenge for businesses is to move traditional legacy systems to the cloud or new data centers, which is not a simple process.
For example, if an application running the enterprise in the 1990s VAX small computer, and will transfer to another traditional data center, then the method of upgrade and transfer "is a feasible solution. However, if you move your application to the cloud, you'll need to migrate it to the VAX emulator before migrating. Otherwise, the enterprise must decide whether to completely overwrite, replace, or delete the application.

Website design a comprehensive comprehensive three steps

Website design, if you are from one industry to another across the industry to understand the unknown in the field of knowledge, it still needs more study, especially now basically every industry needs to use the site, the three steps that the most basic web design also must master the following. Take you to understand in detail.
 Website design
The first step is to look at the same industry website to find its own characteristics and needs
Internet to see our counterparts have site, will have very big help to us. Learn how your peers do business on the web and how to organize content. In addition to peer site release information (product or service information), there may be some good service initiatives, such as collecting customer feedback form, online support, product classification query, even online ordering, online business management etc.. Observing and learning these special features will also inspire the company's decision on how to build a website.
The second step is professional consultation
When the company puts forward the demand that builds a website, may encounter a few unfamiliar terms more often, for instance "domain name", "host", "enterprise mailbox" wait. That's all right. Let's see, the following instructions are clear:
Domain name - sometimes called URL, no matter what international domain name, domestic domain name, is.Com or.Net, for example, there are now.Cc?. They are all names of a website on the internet!
Host - that is the server that receives our web site on the Internet (a dedicated computer). Browser - which software we use to view web pages when we use the Internet?.
"- is our website after the completion of others to watch it to see a picture of the page, which contains you or your unit related pictures and text, a website is composed of many such pages. Among them, there is a web page is more special, is to open the web site to see the first page, it is called home page, also called home page, is the most important web site of a web page.
Virtual host virtual host, as the name suggests, not really rent a belong to our independent host, a place but leased others in a special room real Telecom on the host (host), in the content of the company's website (one page), then I have registered the "domain" refers to the space
The third step is to determine the content of the website construction
Before preparing to build a website, the company must understand what kind of website it needs and tailor the site according to the company's own needs to be the most useful for the company. If the company pays attention to customer service functions, then the company should set up a membership system; if the company needs online customer service, can use the forum or message board to achieve this function; if only release information requires the establishment of a powerful information release system; if it only needs to be posted on the website of the data and the data is do not often change, companies need to build a static website. To build the site in the early planning for a site to establish whether or not a vital, therefore must be related to the leadership of the company needs to communicate with our website builder, this website builder for enterprises more professional design is more suitable for the enterprise website.
After more than a few master, even if the interlacing can produce satisfactory website, I feel for the above three steps, a good understanding is very helpful, website said the simple is not simple, that is also not difficult, necessary to master professional knowledge, have a good website to be nothing difficult.

2017年9月26日 星期二

Cabling system design considerations

Cabling system interface should have compatibility, to meet the different terminal equipment can use conversion, therefore, design and production in different businesses should adopt international standards, meet between telephone switches, network equipment, and the use of computer support to each other.
Cabling
2. flexibility
In the whole system, the use of advanced jumper management mode, terminal equipment of different types can be connected with the information outlet different, and as long as the frame can be realized by simple jumper terminal equipment to change and shift.
3. application reliability
The quality of products chosen in the computer room must be guaranteed and approved by the international authoritative organization and be subject to rigorous test.
4. investment economy
By improving the compatibility of products, the modular and open product structure is established to reduce the manpower, material and financial investment of daily maintenance, and reduce the operation cost.
Composition of wiring in data room
Data center wiring consists of two parts: Data Center, computer room wiring and support space. Data center room includes main wiring area, horizontal wiring area, regional wiring area and equipment wiring area.
The main wiring area includes a large number of main jumpers and core switching equipment, and is also the central distribution point of the data center structured cabling system.
A main distribution area can provide data support and services for a certain area of one or more data center also provides service and support for multiple data centers in other areas, in addition to office area, operation center and other external support can provide regional service and support.
In the horizontal wiring area, the equipment wiring area is typically equipped with SAN, LAN and KVM switches connected to the equipment wiring area. The level of the distribution area mainly includes the level jumpers in a data center can set the computer room on each floor, the level of the distribution area in each layer is provided with at least one, but when the distance between equipment and level distribution area is far away, by setting up a plurality of horizontal distribution area to achieve the purpose of.
The regional distribution area, is a horizontal wiring system contains optional docking point, in the computer room design, terminal equipment usually adopts cabinet or rack, can move the change area regularly, regional socket connection multiple adjacent equipment, to achieve the level of the distribution area and terminal equipment configuration flexibility requirements.
No cross connection is allowed in the connection of the above devices, and the active device shall not be used in the zone wiring area.
The wiring area, that is allocated to the terminal equipment installation space, which contains more widely, including computer systems, communication equipment, server and storage of blade server and so on, in the frame of connecting hardware or cabinet general fixed regional distribution area in the distribution area of equipment.
Jumper and installation tab
In order to follow the IDC room layout design principles, in the purchase choice when you need to pay attention to the jumper with manufacturers with the brand, so as to ensure the transmission performance, data link line at the same time, for both ends of the RJ45 interface is connected with the frame and network equipment selection, should be in accordance with the six standards.
The following requirements should also be met when labeling the premises with a tag:
1) UTP labeling requirements, material requirements of oil proof, waterproof, anti solvent properties and covered with protective film and transparent strong adaptability, good ethylene better, the bottom can withstand bending, suitable for winding.
The tag has the characteristics of wear resistance, high temperature resistance, friction resistance, waterproof and pollution resistance. The label material requirements for this characteristic must be certified by UL969. The label is mainly used for distribution racks to switch jumpers.
2) identification and information panel label distribution frame requirements, material selection of polyester and its functional properties with UTP labeling requirements, label material must also be certified by UL969, the frame identifier and each panel ID information.
3) fiber labeling requirements, not only meet the functional requirements, but also should satisfy the network maintenance personnel health requirements, in the use of functions, its corrosion resistance, friction resistance, moisture resistance, no environmental pollution on the health requirements, so in the choice of fiber label to meet environmental requirements of HS Ro, mainly for the purpose of distributing frame to the switch of optical fiber.

機房建置精密空調中乾燥劑有什麼作用

機房建置精密空調製冷體系中會有許多有害物質經過各種途徑進入到管路中,常見的有以下幾種:
 機房建置
1、出產或修理進程
這時候製冷劑體系都會對外界敞開,不管時刻長短,都會有水氣,灰塵進入到體系管路中,在焊接進程如果沒有通氮氣維護還有可能產生銅氧氣膜,乃至是焊渣;
2、充注製冷劑或冷凍油
如果這些東西的自身有許多固體雜質和水分,都會帶到體系裏邊並腐蝕體系中的金屬部分;
3、在製冷體系運轉中
在運轉中,製冷劑中含水分時會水解生成酸性物質,對金屬產生腐蝕。製冷劑和潤滑油的混合物能夠溶解銅,並會帶到緊縮機在其金屬表面上構成銅膜,然後影響緊縮機的作業。
4、冷凍油
溫度低到一守時冷凍油會分出石蟆,並且關於環保製冷劑運用的POE油,吸溼性比較強,因而更簡單水解出酸性特質。
以上物質的存在會嚴重威脅或損壞體系中的運動部件及需求緊縮氣體的氣缸和製冷體系的功能,所以都要及時剷除。
枯燥過濾器裏邊起作用的資料首要有
1、金屬過濾網:過濾較大固體顆粒(首要用於R22及運用礦物油等對水分不靈敏的中小型體系中);
2、聚脂過濾網:過濾細小顆粒;
3、玻璃棉:過濾細小顆粒;
4、分子篩:過濾較大顆粒固體雜質,首要是吸收水分,是其他物質的3-4倍,也可吸鄰近酸性物質;
5、活性氧化鋁:吸收水分,首要是酸性物質。
6、活性碳:吸收白臘,不過由於冷凍油只要在溫度低於-35度時纔會分出白臘,所以在中高溫運轉 的體系中不需求用到。
常用機房精密空調枯燥劑特色如下,它們依據運用特色分別做成兩種方式:
1、小球組合型,許多枯燥劑做的小球充填在枯燥過濾器內的一個容器內。
2、硬質濾芯型,經過一系列工藝把枯燥劑做成硬質濾芯,然後裝在枯燥過濾器的固定支架上。
枯燥過濾器的功能於分類:
1、過濾才能
過濾網一般由不鏽鋼,磷青銅或黃銅做成,過濾氣態製冷劑一般用70-100目,過濾液態製冷劑可用60目,但膨脹閥和電磁閥前的過濾網需求選用120-200意圖細濾網。不過現在一般不運用金屬濾網,而直接用枯燥劑構成細小流道,或聚脂濾網,或玻璃棉,既能夠吸溼又能夠過濾,過濾才能可達20um(700目),不過如果運用硬質濾芯的,過濾才能只要40um(380目),由於它是經過多孔滲水來吸附水分,所以不太簡單擋住固體雜質,當製冷劑的流速稍大一點就很簡單把固體雜質沖走,如果添加濾芯密度,這又會添加阻力,使壓力捐失增大。這是它的結構特色決議的。
2、吸溼才能
關於製冷體系內的含水量是越小越好,並且不同的體系在關閉前的狀況不同,抽真空時刻長短,充注製冷劑質量好壞都會有影響,所以關於按規範流程出產的製冷產品很難斷定需求運用多大吸溼才能的枯燥器,因而一般枯燥過濾器出產廠家依照ARI710規範的要求來標稱,如DANFOSS的枯燥過濾器名義吸溼才能是在以下條件測得:(ALCO的也是一樣,僅僅個別參數有差異)。
3、功能性
液管用枯燥過濾器,吸氣管用枯燥過濾器(又叫緊縮機焚燬用枯燥過濾器,),它們都能夠去除可能損壞體系的物質,但重點有所不同。
液管用枯燥過濾器首要維護截止閥,膨脹閥等閥件,以及防止毛細管等節省元件的阻塞,在製冷體系中是規範的裝備。由於這兒流速低,這樣過濾作用好且壓降也小,後面跟着的受維護的閥件也多。在液管枯燥過濾器後面一般是跟着示液鏡。還有一種是在緊縮機焚燬後爲了加強整理受污染的管路的才能而專門運用的枯燥過濾器,它比正常運用的具有更大的吸溼和除酸才能。
機房精密空調吸氣管用枯燥過濾器首要維護緊縮機。吸氣管用枯燥過濾器內行業界應用得還不算多,正式作爲體系裝備的一般有兩種景象:
1、現場裝置的體系:由於現場裝置不同於工廠,現場環境雜亂,潔淨程度差,所以最好能在吸氣管裝置枯燥過濾器。(可是國內現在關於中小型如分體式管道機等現場裝置產品都沒有裝置,靠施工單元自身工藝來確保質量,當然這是對R22的產品,不知環保製冷劑有沒有加強這方面的維護。)
2、大型體系:由於緊縮機的成本比較高,在吸氣管裝一個可拆卸的枯燥過濾器來維護緊縮機是值得的。
吸氣管枯燥過濾器爲了確保壓降儘可能小(吸氣管的壓降對體系功能的影響比較大),所以一般比液管用的大(首要指流轉才能)。由於吸氣管用枯燥過濾器首要是在緊縮機焚燬的情況下運用,所以它能在短時刻內就會吸附到很多的水分和酸性物質或固體雜質,這些東西會使枯燥過濾器的流轉才能下減,然後影響功能,所以在吸氣管枯燥過濾器進出口鄰近一般會各帶有一個接口用來測驗枯燥過濾器的壓力丟失,如果測出的進出口壓力差超越下值時,就要替換枯燥過濾器了。
在枯燥過濾器之後一般要裝置一個帶有溼氣指示器的視液鏡,此刻其指示意義如下:
綠色:製冷劑中無危險溼氣;
黃色:在膨脹閥之前的製冷劑中溼氣含量過高。
氣泡:1)枯燥過濾器壓降過高。2) 無過冷。3) 整個體系中製冷劑缺乏。
當壓降超越以下值時必須替換吸濾器:
空調體系:0.50bar;製冷體系:0.25bar;冷凍體系:0.15bar
裝置機房精密空調裝置時必須使流體方向與過濾器標籤上的箭頭方向一致。枯燥過濾器的方位不限,但必需求記住以下幾點:流向朝下的筆直裝置表明能夠快速排放/清空製冷體系;關於流向朝上的筆直裝置,排放/清空則需求較長的時刻,由於需求將製冷劑排出枯燥過濾器。
枯燥過濾器替換條件
1. 視液鏡指示溼氣含量過高(黃色)。
2. 過濾器的壓降過高(正常作業時視液鏡中呈現氣泡)。
3. 製冷體系替換了首要零件,例如緊縮機。
4. 每次製冷體系被翻開時,例如,替換了膨脹閥中的流口組件。不得重複運用舊的枯燥過濾器。如果在溼氣含量較低的製冷體系中運用或者當其受熱後,則它將會釋放出溼氣。

How to data center migration intelligently

Data center migration shouldn't have been so terrible. To be sure, the process is difficult and urgent, but if the organization is concerned about the consequences of the data center migration project, it may well not be ready and need to continue to improve its planning. The old military adage also applies: "war plans do not keep pace with changes in the battlefield."." Data center migration usually requires more time and resources than expected.
 Data centre migration
According to IDC, a research firm, the organization spends more than 1/4 of its IT budget on data centers, which can easily become major cost centers. The chief information officer wants to reduce the budget, but the reality is that organizations have to assess their data center layout. This usually means that organizations have to consolidate multiple data centers into more efficient configurations, or reposition different data centers into more cost-effective locations, or take both.
Migrating critical business and server areas is potentially destructive to IT assets. Even if the migration itself goes well, the CIO still needs to quickly formulate measurable ways to improve it to cope with this difficult process.
Data center migration shouldn't have been so terrible. To be sure, the process is difficult and urgent, but if the organization is concerned about the consequences of the data center migration project, it may well not be ready and need to continue to improve its planning. The old military adage also applies: "war plans do not keep pace with changes in the battlefield."." Data center migration usually requires more time and resources than expected.
Intelligent migration ensures that the data center converts the first attempt to success. Here are four key points for a successful migration of data centers:
Not dependent on guesswork
One of the worst mistakes for businesses is to overestimate their knowledge of IT infrastructure and its supported technologies, applications, and services. Worryingly, this is one of the most common mistakes. There is no doubt that the lack of new technology and staff turnover make it difficult for organizations to retain data centre expertise.
Although everyone makes the greatest efforts in documenting and reporting, they take away useful knowledge as they are retrained or left behind by the staff. Even if the document does exist, it can be of poor quality or uneven quality. For example, after publishing a new application, the developer fails to document the impact of the change in the policy on the night index. This migration is not only clumsy, but it lacks unified truth data.
The answer given here is careful planning. Spend enough time to correctly complete the initial mapping and discovery phase, which will save time and cost in the long run.
Do according to one's ability
It is tempting to see data center migration as part of a program upgrade and to accomplish other major changes at the same time. In theory, it saves time and cost, but in practice it increases the chances of project failure and the rest of the crash. When there are multiple projects, consideration should be taken in order to ensure adequate time for migration.
Similarly, setting unrealistic goals, schedules and doing work without understanding the timing of the schedule is a common mistake. This may begin with a statement, for example, the data center lease expires within three months, so we must migrate to the new data center before maturity". This may not be a problem for a small company with only 10-50 servers, but for larger companies, it will take more than three months. In addition, its complexity also affects time scales. A balanced service delivery and migration deadline will have a major conflict when the same 10-50 servers are bound by the rigid service class agreement (SLA) for all day revenue generating websites.
Even if there is a sensible migration plan, enough manpower is needed to make sure that the deadlines are not missed and that employees are not overly stressed. Assuming that IT support and operators have to migrate the data center's resources and continue to perform their normal duties, this will inevitably reduce the quality of the two jobs.

Website design detailed steps introduced

Access to website design is equivalent to entering a virtual world. Here you can use the mouse to access websites around the world by clicking on links, but that means you're free in this virtual world, and you need a web site to introduce yourself. For businesses, its home page is equivalent to its facade, and the government can build its public image through its Web site. So what specific steps do you need to build a web site? A good and bad site, you can also first from the domain name as a reference, but the quality of the site is not only related to the domain name.
 Website design
One, register domain name and hire server space
Popular, domain name is a name for the web site, and simple, short domain name is more convenient to remember, a good name domain name to the website propaganda is very helpful. Usually promotion website will consider using product name as domain name, that is, product name is also the domain name of the website. For formal units, such as enterprises and institutions, of course, must register their own domain name, on the one hand, if they are not registered in time and others will be registered will cause great trouble, on the other hand, domain name registration fee is not high. Renting the server space is the best choice for the construction of small and medium-sized websites at present. The advantage is that the cost is low, the providers have perfect after-sale protection, and the small and medium-sized websites need not be equipped with professional and technical personnel.
Two, determine the site type
Each site must have its own value, the site should first clear their own goals and types of stations, do not blindly follow the trend, in order to build a site, or else it will be a site that no one visited. Site types include: 1, product (service) query, display type website brand promotion website, 2, 3, 4, foreign business websites, online shopping website, 5, the enterprise portal information website, 6, industry association, information portal /B2B trading service website 7, communication platform, 8 the government information portal website. Determine the type of site, you must determine the theme of the site, the theme of the site must be bright, so as to impress users.
Three. Collecting information
This is a vital part of the website construction. The richness of the data can make a website more attractive, thereby attracting visitors to browse and increasing click through rate. You can find the information you need on the Internet, such as picture scanning, text input and so on.
Four, planning website
After you design the site structure and collect enough resources, you can start the layout page. Plan to set up the page hierarchy of the web site and put the collected files and data in the corresponding directory.
Five edit the document and super connection
After you insert the page layout, the text is of the same type, the same style, the same color, and the picture is large and small. So it is necessary to edit all kinds of elements, for example, change the text font, font size, color, size, etc., the size of the picture, the adjustment of the table, the adjustment of the button behavior, and so on!
Connection is the soul of web pages. When browsing through the web page, click the hyperlink on the web page to jump to the relevant page, so a good web page can not be separated from the connection.
Six, making web pages
Using the software tools we mentioned earlier to create web pages, relatively speaking, Frontpage and Dreamweaver are the most popular tools for making web sites in the internet. After the completion of the production, we have to test the production of the page on the machine, testing content, including page display, links and so on.
Web art design should be consistent with the overall image of the enterprise, to comply with the CI specification. Pay attention to the color of the webpage, the application of the picture and layout plan, keep the whole consistency of the webpage.
Seven, online publishing
Send a good web page to the web and test it. The next job is to promote the website, which is the most important part. Before uploading, you must check your domain name and home page space, as well as page and site connections. Upload software generally use FTP, upload to the server under the domain name.
Eight, later maintenance and propaganda
After the site is released, but also often to site maintenance and publicity. Site maintenance refers to constantly optimize the website function and update web content, site maintenance is also a kind of website propaganda method. For the new online website, publicity is very important, the propaganda is divided into: the search engine PPC, the site itself SEO optimization, new media such as the gate website or related forums, and traditional media such as newspapers, TV and so on.

The measures and methods of data center Containment

The design of the data center Containment does not theoretically fail, but it does happen, so data center operators will face a very severe situation, especially with data centers.
 Containment
According to recent events, the consequences of power outages and business disruptions in the managed data center are very serious. For example: British Telecom is one of the world's largest telecommunications operator and managed data center, the data center operations this year suffered two outages. Voice and data traffic in London and its surrounding areas declined by 10% and the accident time was more than four hours, according to reports.
Despite efforts to avoid interruptions or accidents when designing and running data centers, data center hosting facilities do not avoid these problems, and their short and long term unexpected disruptions can be costly. If the customer chooses to give up the service, the enterprise may be punished for not meeting the service level agreement (SLA), and may cause long-term damage to the brand of the enterprise and cause loss to the business income.
From the data center point of view, what should be done or should not be done to prevent the occurrence of accidents is a very simple truth. However, this is a different conclusion if the data owner and its data center solutions fail. If the enterprise customer has made a strategic decision, put its data in the external data center and carry out risk analysis. But is that really the best way to prepare for the worst? The question is, what should an enterprise customer do if he finds himself in such a situation?
The best way to prepare for the worst is to constantly resolve the possibility. Failure, organizational effort, preparation, and understanding of the process will provide them with resources and tools to mitigate failure. If the enterprise has not considered or did not do so, then it is recommended to judge its own situation from the following aspects.
(1) risk diversification
First of all, when the enterprise makes data center strategy, it should avoid putting all the data in one place, which will increase the risk factors. In the same way, avoid putting all key applications in the same place. Consider storing the primary data in one location and storing the backup data in another location. Then, step by step to understand each scene and determine what impact any level of failure will have. Repeat the process once a year.
(2) trust, but to verify
Businesses get audit records from service providers and, more importantly, careful reviews. In many cases, the hosting data center needs to be audited for compliance with HIPAA, SOX, and PCI requirements. Sometimes, however, this review may be done by people who do not fully understand how IT or data centers operate. Therefore, companies need to arrange for professionals who know how reliable data centers operate. These third party audits are usually much easier than their own identifying risks and can provide more information. In most cases, the cost of mitigating risks is usually minimal through review and validation measures, compared with the costs and operational costs of the disruption incidents.
(3) sign a written agreement
The enterprise needs to know how the data center hosting provider will deal with an outage. When signing a contract with a supplier, a written agreement shall be signed to acknowledge that the agreement will lead to an interruption. This is critical. In fact, data owners find that sometimes protocols do not cover their ideas. In addition, a written assurance shall be made of the service provided by the supplier during the interruption and will be committed within an acceptable period of time.
(4) backup strategy
Companies must understand the risks faced by their business and prepare for the worst. Most managed data centers have an alternative site that can handle basic disaster recovery to ensure that their customers have little impact on operations. Most companies are still seeking to deploy dual live databases in data centers (hosted data centers, cloud computing, or internal deployments). While some of the dual deployments were successful, it was painful to try using a disaster recovery backup. The database is not as complete as the enterprise expects, and the data loss or application is likely to be affected during the failure conversion.

2017年9月25日 星期一

如何避免機房建置中的問題

在過去幾年裏,機房建置發展很快,新建項目很多,也隨之暴露出不少問題:未能滿足企業業務需求,或者在測試驗證或層認證過程中才發現問題,但亡羊補牢的修復需要花費更高,更耗時。這些失誤,無論是數據中心業主還是設計者,都需要承擔一定程度的責任。
 機房建置
專家對大量的案例研究發現,數據中心故障,項目延期,超預算等問題,起因大都根源於建設階段,甚至項目早期的資金階段,正在確定設計目標,預算,時間進度,需求建議書和採購文件發佈,搭建建設團隊的時候。原因是各利益方的目標未能協同一致,以及承包商的表現不良,項目計劃不足和團隊溝通問題。
建造監管,計劃和預算的疏忽,意味着新建數據中心將不能滿足業主對彈性的期望值,或者需要額外時間或者預算,來解決驗證期間甚至更晚些時候暴露的愈發顯著的問題。
導致數據中心表現不佳的建設問題可以歸結如下:
——複雜系統的整合不佳
- - - - - -缺少完整徹底的驗證,或是壓縮了驗證週期
——設計變更
——材料或產品的替代
這些問題出現在建設階段,驗證甚至運維階段,可能影響預算,項目進度或它運營。這些建設問題經常發生是因爲變更管理流程不佳,項目團隊缺少經驗,項目參與者的目標不一致,或者缺少第三方驗證。
1,指定數據中心業主代表
所有利益方在項目伊始就應該認識到,業主的目標大大有別於建造者的目標。業主想要一個能夠最佳滿足成本,建設週期和所有業務需求包括數據中心可用性的數據中心。建造者想要滿足項目預算和週期要求,同時獲得一定項目利潤。數據中心可用性和運營考慮通常在建造者的考慮範圍和專業知識領域之外。
因此,當需要利用承包商在材料和人工成本,設備可用性,本地行爲準則和實踐等方面的經驗,以便節省資金和幫助業主按照既定的項目週期進行建設,而不希望影響可靠性和可用性方面的時候,由項目業主或業主代表全力主導合同語言,流程,權限,以限制承包商去變更或者破壞設計決策的能力,就勢在必行了。
業主應當指定有經驗的代表,審覈承包商的資格,經驗,人的手,領導力和溝通能力。缺少經驗和報價便宜的承包商經常會出現質量控制問題和設計上的偷樑換柱。
業主或業主代表應該完成所有的項目要求,建立大家一致認同,基於運營序列和合理的,有激勵性的建造進度表,包括留有足夠的時間去做嚴格的和完整的驗證。此外,業主代表應該經常性的檢查項目進度,通知團隊成員項目進展,以確保分配給測試和驗收的時間沒有被減少。
項目經理,或承包商,要關注可能要執行序列外任務的進度表。執行序列外的任務以及返工去糾正錯誤放置的電氣服務,管道,線纜等,這些都可能導致項目進度延誤問題。
建造進度延誤不應該用放棄測試驗證來彌補滯後的時間差。在建造合同中加入多項針對延誤的懲罰條款是一個應該考慮進去的解決方案。

The past, present and future of data center migration

Data center migration is the center of modern software technology and plays a key role in expanding enterprise capability. The data center concept originated in the late 1950s, when American Airlines and IBM cooperation, to create a company belonging to the United States Sabre reservation system for passengers, so that this part of the main commercial areas become automatic. In 1960, the concept of a data processing system to become a reality, it is used to create and manage aircraft booking system, so that any agency anywhere can timely access to electronic data, then open the door to enterprise class data center.
 Data centre
Since then, physical and technological changes in the field of computing and data storage have taken us to a winding road to today. Let's take a brief look at the history of data centers, from the previous mainframe to the current cloud centric, and their impact on IT decisions.
1946
Electronic digital integrator (ENIAC) was invented in 1946 for the U. S. Army's storage artillery firing code. It was called the first universal electronic digital computer.
1971
Intel adopted its 4004 processor and became the first universal programmable processor in the market. It provides a "block type" service that allows engineers to purchase and customize software, enabling many electronic devices to perform different functions.
1980
The generation of personal computers (PC), which began in the 1981, led to a boom in the micro computer industry.
Sun Microsystems has developed a network file system protocol that allows users of computer clients to access files over the network in a manner similar to accessing local storage.
Computers are quickly installed on any one of us, but little attention has been paid to the environment and operational requirements.
1990s
Early microcomputers began to serve as servers instead of older mainframes, which became increasingly known as data centers. The enterprise starts to organize the platoon server group in the interior.
1990s
Metaphase". The com wave has prompted the company's desire for fast network connectivity and uninterrupted operation. The enterprise starts to construct the server space which can run more equipment (tens of thousands of servers). The data center, as a service model, became popular during this period.
Note: thanks to the PC (server), the IT decision begins in two separate ways. The server allows for application based decisions, while hardware (Data Center) decisions are still within the enterprise scope.
1997
The Apple Corp created a program called Virtual PC and sold it through Connectix. Virtual PC, like SoftPC, allows users to run window copies on a Mac computer to resolve software incompatibility issues.
1999
VMware began selling VMware PC like Virtual Workstation., the original version can only run on Windows systems and later support other operating systems.
Salesforce.com pioneered the concept of delivering enterprise applications through a simple web site.
2001
VMware ESX release, this is a bare metal management procedures, can be run directly on the server hardware, the underlying operating system without additional.
2002
Amazon AWS has begun developing a cloud based service that includes storage, computing, and artificial intelligence implemented through the Amazon Mechanical Turk.
2006
Amazon AWS has begun offering IT infrastructure services to businesses in the form of Web services, and is now commonly referred to as cloud computing".
2007
Sun Microsystems uses modular data centers to change the basic economics of enterprise computing.
2011
Facebook launched the open computing project, which advocates industry wide sharing of technical parameters and practical experience to create the most energy-efficient and economical data center.
2012
Survey shows that 38% of enterprises have been using the cloud, 28% of enterprises plan to start using or expanding cloud.
2013
Telcordia has released General requirements for telecom data center equipment and space. The document provides minimum space and environmental requirements for data center devices and spaces.
Google invested $7 billion 350 million in huge sums of money in 2013 to build its network infrastructure. The cost is for the massive expansion of Google's global data center network, possibly the largest construction effort in the data center industry.
Today and in the future, data centers are shifting from the ownership model of infrastructure, hardware and software to another on-demand subscription and delivery model.
To meet the needs of the application, especially through cloud computing, today's data center capabilities need to match the cloud. With the integration of data centers, cost control and cloud support, the entire data center industry is now changing. Cloud computing matches today's data centers, making IT decisions more flexible access to resources, but the data center itself is a complete and independent entity.

Design ideas of website design

In web design ways, thought the colourful, cool, can attract people, but wrong, now this day and age, people don't like too complicated things, are now pursuing a minimalist style, simple and comfortable manufacture web page if it is too complicated, will only make a person dazzling, unable to jump out in the total number of peers.
 Website design
Now let's share a few thoughts on how the company's website design is deeply loved by customers.
1. Focus on behavioral call design
Every design on the site should be purposeful to meet the user's behavior or needs, which is obvious.
This also provides the basis for the design of behavioral calls.
Guide the user's copy, the button and the link should be clear, obviously, never let the user miss.
2. Simplified paging
Do you have a page that requires a lot of different pages?
Do these pages need to be subdivided?
You can reconsider the question.
The content of multiple pages is reduced to manageable chunks. Don't let the user click on the content to view the content. The relevant content will be saved together so that the user can focus on reading and viewing.
Delete expired, outdated content and information, and small plug-ins and third-party information are merged into the appropriate chunks, not scattered.
3. Unified color scheme
While many colors are really attractive, they can also be overwhelming.
The color scheme is controlled by two or three colors, which allows you to control the design colors as much as possible.
If you want your color as simple as possible, so you can use a single color, you will find the simple color is so beautiful, the color of the less for the user's psychological load is small, especially when your whole visual design is full of harmony.
4. Use standard navigation design
While hidden navigation looks cool, not traditional creative navigation is also popular, these navigational modes are not so user-friendly.
Deviation from conventional design is easy to remember, but it is often difficult to use experience.
Select the standard navigation mode to make your site easier to use.
The most traditional navigation mode is to place the navigation bar at the top of the page, usually three to eight different options, don't choose a decade ago, of course, popular complex, comprehensive large navigation menu, unless you are so big taobao taobao retail electricity business.
5. Adopt the 80/20 principle
Remember, you will trigger the 20% to 80% of the content of user operations, this means that you should give priority to, the content of the interface elements (such as behavior calls and call button) will become an important component of the direct user interaction.
Choose fonts carefully
Fonts and layouts also follow fewer than many principles.
The easiest font to read, usually with a standard look, uniform strokes, plain and without extra decoration.
A complete set of fonts usually has various word weights and optional styles, and you don't need to look up other styles.
In the layout of the matching font and problem, the text needs to be completely contrasted with the background

2017年9月24日 星期日

The big problem facing server rack

The big problem faced by enterprises in selecting server rack.
1. Increased availability of high availability and mobile connectivity
More devices need access to the network than in the past. A surge in data has led to an increase in rack power density.
Server Rack
Two, IT development needs to synchronize with enterprise development
ACP rack and cabinet products are fast deployed. The ordering of new IT devices (a part number including a fully assembled cabinet and all its components), delivery and setting, installation and deployment, has been the biggest simplification. Because these cabinets are designed by vendor neutral, they guarantee full compliance, avoid a lot of installation problems, and ensure that all deployed rack mounted devices can be properly installed in the cabinet.
Three, on-site IT manpower shortage
The ACP rack has standard functionality, easy to use, and efficient, streamlined deployment and installation procedures. The pre installed leveling foot and casters also omit the need to order and install such components. Simple and adjustable vertical mounting rail makes the equipment easy to install. These outstanding features can help SMEs save valuable time.
Four, the diversity of IT deployment environments
The ACP rack is suitable for a wide variety of applications - from an open rack, wall mounted equipment, to a lateral airflow rack with a network switch to meet the requirements of a variety of mission critical IT facilities.
Five, optimize management to meet the changing needs of IT
All rack type ACP offered by the PDU can be correctly mounted in the rack. Customers can choose to upgrade the basic rack type PDU configuration to switch type rack PDU as needed, and get more perfect remote device management capabilities.
Six, the increasing complexity of IT
Trends such as virtualization and integration have prompted the need for different types of IT devices in multiple racks. ACP provides a wide range of rack and rack products to meet this trend and provides an efficient and flexible application environment for a wide range of different devices.
Seven, use wisdom to manage energy
The ACP metering rack type PDU monitors power and energy consumption at the rack level, provides data to the IT manager in a timely manner and makes informed decisions to ensure maximum efficiency for each rack.

機房建置性能要求

A級機房建置的基礎設施宜按容錯系統配置,在電子信息系統運行期間,基礎設施應在一次意外事故後或單系統設備維護或檢修時仍能保證電子信息系統正常運行。
機房建置
A級機房建置涵蓋B級和C級數據中心的性能要求,且比B級和C級數據中心的性能要求更高。
意外事故包括操作失誤、設備故障、正常電源中斷等,一般按照發生一次意外事故做設計,不考慮多個意外事故同時發生。
設備維護或檢修也只考慮同時維修一個系統的設備,不考慮多系統的設備同時維修。
在一次意外事故發生後或單系統設備維護或檢修時,基礎設施能夠滿足電子信息設備基本運行需求。
A級機房建置同時滿足下列要求時,電子信息設備的供電電源可採用不間斷電源和市電電源相結合的供電方式。
設備或線路維護時,應保證電子信息設備正常運行;
市電直接供電的電源質量應滿足電子信息設備正常運行的要求;
市電接入處的功率因數應符合當地供電部門的要求;
柴油發電機系統應能夠承受容性負載的影響;
向電網注入的總電流諧波含量不應超過 10%.
在保證可用性的前提下,降低數據中心總體擁有成本(TCO)。
電子信息設備屬於容性負載,柴油發電機系統應能夠承擔容性負載的影響;
當電子信息設備產生的電流諧波超過 10%時,應進行諧波治理。
當兩個或兩個以上地處不同區域的數據中心同時建設,互爲備份,且數據實時傳輸、業務滿足連續性要求時,數據中心的基礎設施宜按容錯系統配置,也可按冗餘系統配置。
這是A級機房建置的一種情況,主要適用於雲計算數據中心、互聯網數據中心等。
當兩個或兩個以上在同城或異地同時建立的數據中心互爲備份,且數據實時傳輸備份、業務滿足連續性要求時,由於數據中心之間已實現容錯功能,因此其基礎設施可根據實際情況,按容錯或冗餘系統進行配置。
B級數據中心的基礎設施應按冗餘要求配置,在電子信息系統運行期間,基礎設施在冗餘能力範圍內,不應因設備故障而導致電子信息系統運行中斷。
B級數據中心涵蓋C級數據中心的性能要求,且比C級數據中心的性能要求更高。
C級數據中心的基礎設施應按基本需求配置,在基礎設施正常運行情況下,應保證電子信息系統運行不中斷。

The vast project of data center migration

Data center migration is a huge project with great cost. We usually say this about the data center, data center, in fact, no matter what, the construction of the data center to go through a long process, some a few years or even decades, after such a long time to be put into operation. We often talk about all kinds of rapid deployment of data center technology, which is the data center big frame have been playing, just connect all kinds of IT equipment, the process of debugging normal working, then more is the software level adjustment, before a part, and this is just the construction of data center here, to do a lot of preparatory work for the construction of a data center, we have to talk about.
Data centre migration
First of all, the data center migration need to build some equipment room environment, which of course need land and get permission from government departments to start construction. The data center location itself is a long process, the construction of the enterprise data center to inspect various parts of environment, transportation, personnel, hydrological and geological and natural disasters in advance, of course, there are various places for preferential policies, it is necessary to consider the important aspects of the. In some places in order to create jobs, in order to promote local economy, enhance the level of information of local, by providing cheap land use fees, municipal facilities costs, utilities costs and so on to attract people to the local data center construction, which can save the initial investment and operation cost for large enterprises. After a variety of visits, judgments, the final data center will be selected to build the location, and then started construction. If there is no local government and relevant departments, fully communicate that city planning, environmental protection, power supply, gas supply, water supply and other departments, is likely to come to naught, not construction.
Secondly, the data center migration construction with reference to the national standard, must be in accordance with the national standards, only after acceptance by the national data center are eligible to put into use, not free to change the architectural design. And ordinary buildings are different, the data center building to load-bearing capacity, construction can not be too high, each floor should be particularly strong, because it is covered with dense equipment. The data center's auxiliary facilities must be complete: overhead floor, hot and cold channel, storage battery, standby power supply, air conditioning pool, grounding system and so on. This is different from the ordinary building. Therefore, the data center architectural design should be experienced by a professional design team to do the design, once the beginning of the design is unreasonable, it will be difficult to modify the later stage. The design of data center should be forward-looking and meet the application requirements for the next ten years, and maintain a certain design redundancy for expansion.
Third, the data center requires uninterrupted operation throughout the year, so as long as two-way power input is needed, continuous power supply is ensured. At the same time, the design of the standby power supply can not be continued forever. Once the power failure occurs, the data center should have the ability to switch to the standby power and continue to operate temporarily. The general use of battery or standby power diesel power mode, continuous power supply time is not too long, or simply the rapid restoration of power supply capacity, demand for data center power is too large, also need to give guidance on the power supply department can meet the power supply of the city, the electricity sector will be evaluated according to the capacity of the data center, if not meet, the data center can be built in this city. There are network lines, the data center if you do not interact with the outside world, the data center is an information island, can not run. Before there are several data center operators for cable Waduan, resulting in business interruption events, visible network data centers on how the key line. The data center is generally used for line backup mode, connect to the network operator through multiple links, as long as not all the fault on the line, that will be all at the same time for cable Waduan, that only through the disaster recovery data center, through the business switching between data centers, to ensure that the business is not affected.
Fourth, the data center infrastructure, as with ordinary buildings, the data center also needs to have the road, gas, sewage, communications, lighting, landscaping, and data center should pay attention to the choice of building materials, must use the sound insulation materials. The data center in the tens of thousands of sets of equipment at the same time, the noise is very high, there may be a pollution to the surrounding environment, especially the need to work in the data center of the people, these people although not in the room, but also in the data center inside the office, they should always maintain the safe operation of the data center. In this way, the data center in each section to consider the problem of noise, the noise in the circle inside the room, not affect the operation and maintenance work, but can not make noise leaked to the outside, has the influence on the surrounding environment, the data center should pay attention to noise in each part of the design, to avoid the noise transmission outside the room.

The importance of website design to enterprise

Enterprise web design, the first impression is always important, usually you only need a few seconds can be impressive. For example, when we first met: dressing, hair style, figure, and so on, we can produce a difference, which brings a delicate special emotion.
Website design
Similarly, first impressions are also important in the online world. Site visitors can be able to judge your website in a matter of seconds, learn about the reputation of your company, start with what they see, and how you feel when you browse your site. According to statistics, if you make a false first impression, 1/5 of visitors will leave your website without giving you a second chance.
Of course, from the beginning to the present, the web design has made great progress over the years. New features are discovered and implemented each year, and the most popular features are added to the list of options used by designers and developers. Unlimited scrolling, landing pages, banners, popups, cookies, social media, integration, sharing, reviews - lists can go on - just a few of the features that have been considered hot trends over the past few years.
Understand the latest web design trends and adjust to deliver the best information so that you can create a website that will attract viewers and attract the attention of the audience.
But before going deeper, let's be clear: even if web design requires creativity, it's not the purest meaning of art. Because beyond emotional expression and passion, web design can help people find answers to their needs, take action or accomplish tasks, and give users a special experience.
More than ever, everything from web pages to content to navigation and layout has a specific purpose. Of course, many websites have been designed for some time and have collected a lot of clutter during this time. If you plan to redesign or redesign a web site, your first step should be to review the content and review it. You should clear it up and focus on the relationship with the target audience. However, if you decide to design a site from scratch, then you should start with site maps, wireframe, and top-level tasks. In this constant, clear and absolutely complete struggle, the first should be the winner.
Beyond all modern web design trends should be the relationship with the target audience and the intention to communicate clear information. So we've chosen five web design trends that will help you create a modern and efficient web site.
1: text typesetting
Even if we are bothered for typesetting, but now there is a narrowing trend to determine the layout, site too many elements, now the expression less and an ornamental image as the background, has now become a.
This type of composition allows, on the one hand, smaller fonts to allow the layout space to breathe; on the other hand, it increases the interest in the information they express.
Balance is also a problem related to this trend. In order to create a sense of harmony, all fonts need to be reduced in some way. This reduction may not only be the size of the font, but also in terms of color. Large and small spelling combinations are very acceptable, where light characters for titles can be used in combination with small fonts from replicas.
2: Geometric hierarchy
In looking for new elements that can add new visual interest to the design, geometric layering is steadily increasing in popularity. Geometric layering is a technique for the freshness of images, such as buildings, interior design, staff, and so on.
Adding interesting shapes, contrasting colors and incisions, always provides a starting point for the user to enter the site design and effectively helps him focus on the most relevant content, such as a brand or title.
The best part of using geometric layering is that they are easy to use and integrate with the rest of the design, and are almost suitable for any type of content. The angles and curves of geometric elements help users move from one element to another, and may be a good orientation tool when they are used effectively. By using the circle for content, you need to first see and fix the angle so that you can point to what you want the user to see, and you will help the audience better understand the message you want to deliver.
3. enhance the stereo feeling
3D was used in Web Design in 2011, and over the next few years, we've seen the possibilities of using 3D among many exciting web designers. The development of 3D in the online world has been going on, with some very impressive implementations from visual effects as proof of concept, and the formal use of 3D, designed to make a great network experience.
Using 3D elements in web design can make your site more memorable and memorable, and its purpose is to create a sense of depth. As a result, sites including 3D elements are more interesting and dynamic than most web sites that contain graphic 2D designs. In terms of the use possibilities, the 3D effect can be well applied to a variety of design options. From the game style interface to shopping and product identification, the interface to virtually interactive user authentication is quite attractive. This may be the main reason for integrating 3D effects and styles into web design. In addition, with the rapid development of VR / AR, this trend is developing rapidly and will be more popular than it is now.

2017年9月22日 星期五

Computer room data center fire system

Fire system:data center, room IT system operation and storage are the core data, the IT equipment and other equipment related to the special requirements of the fire itself, must be on these important equipment design of fire protection system, and relationship between IT and the normal operation of equipment to protect the equipment off health; room fire extinguishing system, prohibiting the use of water foam and powder extinguishing agent, suitable for gas fire extinguishing system; fire engine system should be relatively independent of the system, but must be linked with the fire center. Large and medium-sized computer room, in order to ensure the safety and correct grasp of abnormal state, once a fire can be accurate and rapid alarm and fire, automatic fire extinguishing system is required.
 Fire system
Design process of gas fire extinguishing system:
According to the design specification, determine the room where the gas fire extinguishing system should be set, and select the type of gas extinguisher.
Divide the protection area and protection space, select the system form and confirm the location between the bottles.
According to the relevant design specifications, the design amount of fire protection area is calculated, and the number of fire extinguishing bottle is determined.
Determine the layout of the bottle in the storage room, check the size of the bottle storage is appropriate.
Calculation of protection zone fire extinguishing agent conveying the average flow of the main way, diameter and number of nozzles set the competent road.
According to the actual situation of protection zone interval uniform layout sprinklers and pipes to try to set up a balanced system, the initial pipe diameter.
According to the calculation method of pipe network in design specification, the layout of pipe network and the pipe diameter of each pipe are checked and corrected, and the specifications of each nozzle are determined until the specifications are met.
According to the design plan, statistics system equipment materials.
Comprehensive evaluation of design plan and optimization adjustment when necessary.
Automatic fire alarm system

數據中心機房建置中的設計配合

數據中心的核心就是機房建置。爲了保證設備穩定可靠運行,機房必須滿足機房內設備以及工作人員對溫度、溼度、潔淨度、風速度、電磁場強度、電源質量、噪聲、照明、振動、防火、防盜、防雷、屏蔽和接地等要求。先進合理的機房工程建設越來越重要。
機房建置
但是,在機房建設的過程中,由於受到規劃、建設單位、建築條件以及各個專業的現場條件限制,設計出來的機房往往出現空間不夠、機房高度不足、空調規劃不合理、機房柱網不合理以及經費不足等;如果是由舊樓改造成的機房,還會面臨承重不足、加固方案不合理以及原建築的給排水環境等問題。
因此,我們在進行機房工程建設的過程中,需要面臨與建設單位、設計單位的建築、結構以及水暖電專業配合上的問題。本文將針對以上問題進行闡述。
1與建設單位的配合問題
1.1 機房等級的確定
在我國規範GB 50174—2008《電子信息系統機房設計規範》中,根據機房的使用性質、管理要求及其在經濟和社會中的重要性將機房分爲ABC三級機房。每一級的機房都對機房的穩定性有一定的提升。其中,A級機房應能達到容錯的標準;B級機房對機房重要的基礎設備有冗餘的要求;最經濟的則是C級機房。
但是,現在很多的設計單位並沒有對建設單位提出的機房等級進行引導,而且建設單位在建設前期,一般不會很好地定位他們的機房等級,結果就是出現一味地提出超高等級的機房要求的情況。隨後往往又因爲投資過大導致機房沒有辦法按照設計成果進行建設,然後反覆讓設計單位修改圖紙,增加工作量。
因此,在項目建設的前期和建設單位溝通時,設計單位應當對需要設計的機房的各種因素有一定的認識,並清楚該機房應有的機房等級。在前期與建設單位溝通的過程中,結合項目的特點和投資等因素引導建設單位進行正確的定級,使得設計的機房更符合使用的要求。
1.2 IT 設備的設置需求
數據中心機房建設的目的,是爲IT設備的穩定運行提供良好的環境。其中就包括機房環境、用電環境以及周圍環境等。因此,IT設備是數據中心機房的核心。
但是,在以往的建設過程中,一般的建設流程爲:選址→設計→裝修→設備→IT設備。可見,IT設備的設置及運行被放到了最後。其原因是受制於建築外觀、建築其他功能等原因,機房設計一般都會在最後進行。機房IT設備開始進行設計時,所有的建築外觀、機房面積、建築物其他功能都已確定。因此,機房的設計及實施就受到了很大的限制,容易造成如下結果:
1)機房面積不足、機房的尺寸、柱網規劃不合理、承重不滿足要求等,造成面積大量的被浪費、機房空調不能合理規劃、電源室位置不理想且面積不足等後果。
2)因爲沒有提前考慮用電問題,導致機房供電容量不足,不能滿足IT設備運行的問題。
3)因爲建築位置及內部佈局已確定,導致機房的周邊環境以及機房內部專用的水管、消防管等不能很好地隔離,留下一定的安全隱患。
因爲機房的建設主要是爲了讓IT設備能更穩定地運行,因此,在這裏提出,機房的正確的建設流程應當爲:IT設備→設計→設備→選址 →裝修。
我們首先確定了IT設備需要的容量、運行方式及後期的發展預留等內容,才能很好地規劃出需要建設的數據中心的面積、用電量、製冷量以及後續的操作區域等。在這些內容都確定後,再去規劃需要建設的地址和內部的裝修設計,纔可以建設出完全符合IT設備穩定運行需求的數據中心機房。
1.3 投資規劃
機房的投資也是影響機房建設的1個重要因素。很多部門要求建設數據中心機房,但是因爲前期的投資規劃不合理,最終導致建設不完整甚至出現被迫下調機房等級的結果。
因此,我們在和建設單位進行投資規劃的時候,首先應當瞭解建設單位對IT設備運行的整體需求。在確定機房等級並同時考慮後期發展的預留空間後,再推算出需要的用電量、製冷量、操作區域、設置系統等其他因素,最終得到可以滿足建設單位IT設備穩定運行的機房的投資。