Huawei cloud computing help Sichuan Telecom ICT business development Client background

With the SCUN rapid development of the global environment, the way people communicate more and more diverse, more and more fast development of information technology, the Internet bring new wealth of business experience,it puts forward new requirements for Telecom operators. The telecommunication business development needs of calculation, by a singlevoice and data service cloud intelligent pipeline and the mobile Internet, to consumers by users to pay more attention to enterprise users to provide theICT service integration.

Under this background, Sichuan Telecom in 2010 July and Huawei signed astrategic cooperation agreement with cloud computing, and both leaders met on several occasions, in-depth study of the short-term and long-term businesscommercial planning, and the establishment of the joint development and joint planning group, the comprehensive construction of Telecom cloud computing ecosystem.

Challenges facing customers

Sichuan Telecom Service Platform Facing maintenance and construction andmany challenges, mainly:

The utilization rate of resources are uneven: physical resources in existing data center utilization rate is low, high energy consumption, high cost of building new business;

Business platform: has built more than 100 platform service network platform,business platform independent, long cycle of new business online;

Platform: platform decentralized management of distributed data management,distributed storage, maintenance time is long.

Solution

In 2011, Sichuan Telecom. The sea of clouds, the cloud management platformHuawei virtualization and unified, construction of cloud computing data center,at the same time integration of the existing IT infrastructure, cloud computing resources pool formation of internal migration, integration service platform and the development of cloud desktop business, foreign development of Cloud Citycloud hosting service.

As of 2013 March, Sichuan Telecom cloud computing platform issued a total of 600 multiple virtual hosts, bearing the number 118114, the digital home andvehicle location and other 20 kinds of business platform and cloud hostingservice. Sichuan Telecom Construction cloud mall, the user through the self-help for cloud services portal, be self-help application cloud hosting, has satisfied the enterprise and the end user IT business development needs,convenient for cloud computing services.

Customer value

Cloud computing will bring new opportunity for the development of SichuanTelecom, the sea of clouds effective project integration of IT resources, rapid response to customer needs, and effectively solve the original challenge:

Use of the server rate of 15% to 85%;

To improve the business line speed, single business online time is shortened by 80%;

To realize the resource intensive sharing, achieve the purpose of energy saving;

Create a cloud computing industry chain, using cloud computing platform and the third party cooperation, SCUL bearing industry application.

more information ,you can visit http://www.3anetwork.com/cisco-catalyst-2960-switches-price_c39

Huawei CloudEngine data center exchange conference held

The world's leading ICT MA5680T solutions provider Huawei in Beijing Kerry Center"cloud engine, bearing the future -- CloudEngine data center switches andCloud Fabric solutions conference", to the guests was to promote the coreswitch CloudEngine series data for cloud computing. The release, in addition to the virtual network of CloudEngine data center switches can provide singlebox 48T the industry's largest exchange capacity, high density 100GE/40GE/10GE/GE port, comprehensive, computing, storage, network convergence properties, Huawei has released a new report about Americaauthority testing laboratory of Miercom, further verify the huge advantageCloudEngine data center switches in the forwarding performance and green energy, is the construction of network core equipment very large, low energy consumption data center is preferred, data center level switch products of be worthy of the name of the future.

The conference, Huawei business BG CEO Xu Wenwei, member of the expert committee, national e-government e-government industry renowned experts,the state information center, former chief engineer Ning Jiajun speech,research vice president and Deputy China area president Huo Jinjie sharedthe "trend of data center network and the challenge of IDC Greater China region". Huawei enterprise network chief architect Chang Yue brings "Cloud Architecture Architecture -- Huawei Cloud Fabric data center network solutionanalysis," Huawei enterprise business solution sales CTO Wu Yisheng "why is cloud engine -- Huawei CloudEngine data center switches introduced", thecharacteristics and advantages of Huawei Company for next generation datacenter "the cloud" High Performance Switches and solution, vivid display in front of the guests.

Conference of the product display area, in addition to display CloudEnging128/68/58 series switches, Huawei Technology experts and connect the tester through CE12808, CE6800 electrical interconnection, successfully simulates the flow test environment, to attract a lot of customers to observe,communicate. In the display area, Huawei also sync released in North Americaauthoritative testing laboratory of Miercom on CloudEngine data centerswitches evaluation report. Data of the evaluation report shows, CloudEngineseries of products in the forwarding performance and green energy-savingaspects of the performance is particularly prominent, CE12800 single Qatar1T forwarding capability, support high density 24*40GE interface card full wire-speed forwarding, can fully meet the data center network of large capacity,non-blocking exchange demand; CE6800 hardware architecture of advanced design, 10GE port access the industry's highest density, store-and-forwarddelay is less than 2 μ s, much lower than the industry average level; at the same time, the CloudEngine series have a good performance in energy consumption test, fully embodies the design concept of CloudEngine series switch green energy-saving

Conference opening, Huawei first reviewed the CloudEngine data centerswitches and Cloud Fabric solutions world tour, from North Las Vegas Interop show "unveiled" to Western Europe, London Ovum CIO conference was"enthusiasm", eventually returning to China Beijing's independent release,realize the world tour the perfect ending. Huawei business BG CEO Xu Wenwei in his speech, further express the Huawei attaches great importance to theChinese market with respect to China, since customers long-term support andconcern for.

IDC Greater China region, research vice president and acting president of China, Huo Jinjie said, research shows that we are entering a new era, a newIT platform. Social networks and the rise of cloud computing, mobility and large data flow demand, growth will bring the demand of data center traffic. How tolet the data center network has the greater capacity, faster connection speedand higher security, is the three issue of the CIO were most concerned about the.

Huawei enterprise network chief architect Chang Yue from the elastic (Elastic),ease of use (Easy-to-use) and open (Open) three aspects, strategic visionexplains Huawei Cloud Fabric data center solutions, vividly demonstrates theHuawei cloud network architecture to architecture. Huawei enterprise business solution sales CTO Wu Yisheng showed Huawei Cloud Fabric solutionpanorama, and focus on the interpretation of the three core advantages ofHuawei CloudEngine series -- data center "next generation cloud" high performance switch, namely elastic cloud network (Scalable Fabric), virtual cloud network (Virtualized Fabric) and MA5683T fusion cloud network (ConvergedFabric), demonstrates why the customer to choose the Huawei cloud engine --CloudEngine series switches.

more information ,you can visit  http://www.huahardware.com/

MPLS fault: Tracert switches do not display a middle path

If the CE1 tracert CE2, SCUN can show the middle way, on the contrary, if the CE2tracert CE1, the first jump to NE80E, second to CE1, no VPN pathways of the intermediate display.

Fault analysis

Because the user network public network private network is introduced, also can pass directly from the tracert NE40-4 private network, so in the CE1tracert CE2 can show the public network address, this is up to the public.

By default, the VPN message is not enable IP TTL replication function, so thetracert on the CE2 CE1 show only private network path

- end

Processing steps

Perform the following operations on NE80E.

Step 1 executes the command System-View, enter the system view.

Step 2 executes the command MPLS, go to the MPLS view.

Step 3 executes the command TTL propagate public, enabling MPLS public network message IP TTL replication.

Step 4 executes the command TTL propagate VPN, enabling MPLS VPN message IP TTL replication.

Return to step 5 executes the command return to user view, execute the command save, save the configuration changes.

- end

After completion of the operation, the tracert in the CE2 CE1, can display the middle way. Troubleshooting.

Case summary

Tracert switch MPLS failure does not show the middle path can be solved, to sum up, by default, the VPN message is not enable IP TTL replication, SCUL need toexecute the command TTL propagate VPN to enable this function.

Huawei million desktop cloud container data center by TIER III certification

Huawei MA5680T million desktop cloud container data center TIER III certificationceremony was held at the headquarters of UPTIME COO Julian, Huawei,Kudritzki certification. As the best implementation case of industry TIER III design certification, container data center once again become the focus of the industry.

July 18th news: a leading telecom solutions provider Huawei, recently acquiredTIER III UPTIME design certification issued by UPTIME COO, Julian Kudritzki inHuawei headquarters personally awarded the certificate, 000 desktop cloudcontainer data center (hereinafter referred to as million desktop cloud project)has become the focus of attention of the industry.

Certification on the same day, from UPTIME Network APAC in 2013 will be themembers, including UPTIME COO Julian Kudritzki, North Director Philip Hu,more than 30 people, conducted field visits to million desktop cloud data center. UPTIME member design highlights of million desktop cloud projectgave a high evaluation, recognized the industry best practice, and said: milliondesktop cloud is integrated with the DCIM (Data Center Infrastructure Management), the infrastructure management becomes easy. Secondly, the strength of electrical isolation of deployment, and wiring system specificationcompact easy maintenance. At the same time, the hot weather in Shenzhenand a large number of blade server high-density deployment, pose serious challenges for energy saving, but million desktop cloud box hot channelisolation is reasonable, the proximal energy saving measures of air, theefficiency index of excellent.

Container data center as a kind of new mode, in the rapid deployment,deployment, backup, staging has irreplaceable advantages of mobile fieldscene applications and enterprise park use etc.. In the future, the containerdata center, will become an important mode of the construction and application of power flow. Huawei network energy expert Xiao Bei at the ceremony on thedetails of background of the project: "Huawei million desktop cloud project is located in the headquarters of Shenzhen Huawei, the whole project based on Tier III standard planning: distribution, refrigeration system is designed according to 2N architecture, no single point of failure, can support 10000 of the size of the desktop cloud computing, at the same time the whole 24 hours unattended intelligent operation management. Through the efficient UPS,closed hot channel, efficient air conditioning chilled water row level, single cabinet of 9kW high-density deployment and intelligent management of core technology, is of low PUE value in hot areas, energy efficiency indicators advanced."

At present, Huawei container data center has become one of the fastest growth of similar products, has been large-scale commercial in China, Europe,the Middle East, Asia, Latin America, Africa and other places, covering the field of telecom operators, science and Education Office, medical affairs,mining energy and each kind of enterprise, MA5683T the broad application prospect.

more information ,you can visit http://www.3anetwork.com/cisco-catalyst-2960-switches-price_c39

Five kinds of main technical data center switches

1 high capacity equipment  Huawei Switch

The characteristics of network traffic data center has a high density application scheduling, surging burst buffer, while the ordinary switch to meet the interconnection as the main purpose, unable to realize the accurateidentification and control of the business, to achieve fast response and zeropacket loss can not be in the business situation, business continuity can not guarantee the reliability, the reliability of the system mainly depends on theequipment. So the common switch can not meet the needs of the data center,data center switches must have high capacity transfer characteristics.

Data center switches must support the high-density Gigabit card, namely 48port Gigabit card, in order to make the 48 port Gigabit card can full wire-speedforwarding, data center switches can only be used CLOS distributedarchitecture. In addition, along with the 40G and the popularity of 100G,support the 8 port 40G card and 100G card port 4 also gradually commercial,data center switches 40G, 100G card has already appeared to enter the market, so as to meet the requirements of application of high density data center requirements.

2 cache technology

Data center switches changed the traditional port buffer exchange system, the distributed cache architecture, cache than ordinary switches also big many,more cache capacity can reach 1G, and the general switch can only reach 2~4M.

The sudden flow cache capacity to 200 milliseconds per port in 10g line speedconditions. Thus in the case of bursty traffic, a large cache can still guarantee the network forwarding zero packet loss, is adapted to the data center servercapacity, characteristics of sudden large flow.

3 virtualization technology

Characteristics of data center network equipment requires executives reason and high safety reliability, so the data center switches also need to supportvirtualization, virtualization is the physical resources into logic can be managedresources, to break the barriers between the physical structure.

Virtual network device includes multiple virtual a, a deficiency, deficiency etc..Through virtualization technology, network equipment for unified management of multiple, can be completely isolated on an equipment business, so the data center management costs decreased 40%, the IT to improve the utilization rate of about 25%.

4.FCOE Technology

The traditional data center often has a data network and a storage networkand data center network, a new generation of fusion trend is more and more obvious, the emergence of FCOE technology makes the network integration possible, FCOE is the data frame encapsulation storage network forwardingtechnology in the Ethernet frame. To achieve this fusion technology must be in the data center switches, ordinary switches are generally does not support FCOE function.

5.TRILL Technology

Data center in the construction of two layer network, the original standard STP protocol, but its so some defects such as: STP is to work through the portblocked, all redundant link not to transfer data, resulting in a waste of bandwidth resources;

STP has a spanning tree, the data message has to go through the root bridgeconnecting to arrive, affect the efficiency of the whole network forwarding. So STP will no longer suitable for expansion of large data center, TRILL is due tobe the STP of these defects, is for the data center application and generationtechnology.

The TRILL protocol to the two layer configuration and flexibility and three level fusion and scale efficiency together, the big two layer does not need to configure the circumstances, can realize the whole net loop-free forwarding.TRILL technology is a data center switch layer two basic characteristics, which is not possessed by common switch.

Several kinds of network technology is more than ordinary switches do not have, is the main technical data center switches, is a new generation of data center, data center and cloud service network technology. With these newnetwork technology, Huawei distributor makes the rapid development of data center.

more information ,you can visit  http://www.huanetwork.com

The working principle of a backbone switch

Many new Huawei distributor Client/Server applications and multimedia skills, led to the shared network traditional far can not meet the requirements, it will promote the nextarea network Internet backbone switch show.

1, the definition of Internet backbone switch

LAN Internet backbone switch has many ports, each port has dedicated bandwidth himself, and can connect the different network segment. Between each port Internet backbone switch communication is also, in parallel, which greatly promoted the information throughput. In order to further enhance thefunction, each port can also connected only one device.

In order to complete the Internet backbone interconnection between the switchor high-end server connection, LAN Internet backbone switch common have high-speed port of one or a few, such as 100MB Ethernet port, FDDI port or a 155MBATM port, and to ensure that the entire network transmission function.

The frame buffer Internet backbone switch is mainly from the promotioncohesive server port speed and response of the size, to improve the function of the whole network, and then to meet the requirements of the customers.Some high-end Internet backbone switch adopts full-duplex skills to further enhance the port bandwidth.

The previous network equipment basically using task approach half duplex, i.e.when a host sends a packet, it will not be able to accept packets whenreceiving data packets, time, will not be able to send data packets.

Because of the full duplex skills, namely the host in the transmission data packet also, also can receive the packets, the popular 10M port can lead to 20M port, 100M port can become popular 200M port, it will further enhance theinformation throughput.

Principle 2, the Internet backbone switch

The essence of traditional Internet backbone switch is provided with a multi-port bridge traffic control, namely the traditional (two) Internet backbone switch. The routing technology into Internet backbone switch, can complete thenetwork layer routing, it is known as the three layer exchange, this is the new development of Internet backbone switch.

The Internet backbone switch (the two layer exchange) task truth Internetbackbone switches and bridges, is a task at the link layer, network equipment,each port it has bridging function, each port can connect to a LAN or a highfunctional web site or server, device can through the self learning to investigate each port connection status. All ports are controlled by dedicatedprocessor, and control governance forwarding information via a bus.

Also can use specialized management software for centralized management. In addition, the Internet backbone switch in order to improve the speed and efficiency of data exchange, the general support of various methods.

(1) for storing and forwarding:

All the usual practice of bridge are using this approach. They are the data before frame column other port, the received frame storage memory in the internal, the inspection and then sent to the other port, so the delay is equal to the sum of time and treatment time take a good data frame. If the cascade is long, will lead to serious problems of function, but this method can filter the data frame error.

(2) entry method:

This method only check the data frame destination address, which makes thedata frame Jane when erect can spread, and greatly reduce the delay.

The defects are: frame error will be reported to. Small probability of errorframe conditions, can be used to improve the transmission speed of entrymethod. Huawei Switch The frame error probability larger circumstances, can be used tostore and forward / retransmission to reduce the error frames.

ore information ,you can visit   http://www.huanetwork.com