Enterprise data center "cloud" transformation solution

This article is about 5000 word , About reading 7 minute , Here is the main body .

Digital transformation is having a profound impact on enterprises , In the face of business digitalization 、 The Internet, 、 Intelligence for enterprise data center in performance 、 agility 、 Higher requirements on total cost of ownership, etc , Traditional data centers have become increasingly difficult to cope with . adopt “ Clouding ” Complete the transformation of data center modernization , In order to support business innovation has become a consensus in the industry .

stay 《 Hyper fusion is becoming the building of private clouds “ footstone ” The mainstream solution of , But do you know the relationship ？》 In the article , We have introduced the main components of the private cloud solution , And describes the enterprise data center IaaS “ Clouding ” The importance of transformation , This article will further explore the enterprise data center IaaS Some key elements of upgrading and implementation plan .

How to define an enterprise data center “ Clouding ” success ？

Facing many products and solutions on the market , Many users are in their own enterprise data centers “ Private cloud ” In the process of transformation , They hope to confirm whether their chosen technical route is correct 、 reasonable . In response to this question , Well known authorities Gartner stay  China Summary Translation: ‘Rethink Your Internal Private Cloud’  China has given a lot of valuable views on private cloud construction in the world and in China . The first conclusion is that ：“Many organizations claim to have adopted an internal private cloud. However, the definition suffers from being widely misused. ”（ Many organizations claim to have adopted the private cloud , However, this definition is actually being misused ）, Many of them will use OpenStack Or only completed the virtualization transformation, understood as the realization of private cloud transformation . So ,Gartner In this paper, the following preliminary definitions are given .

The above content is quoted from Gartner “Rethink Your Internal Private Cloud”

actually , Unlike the public cloud, users only need to pay attention to resource usage and payment , Private cloud not only requires scalable resource delivery 、 The ability to be resilient , The bigger challenge comes from ： The overall plan is from planning 、 Deployment to delivery 、 All aspects of operation and maintenance need to meet the following core requirements ：

1. Invest on demand ： We can make on-demand investment at different levels and in different scales according to the actual needs of the business . for example IaaS Layer and the PaaS The phased construction of layers ,IaaS The construction of management platform and resource pool in the layer , And the gradual expansion of the resource pool according to the business situation ;
2. Fast on-line ： Private cloud not only requires rapid delivery of resources , It also requires supporting the automatic deployment of business systems and convenient online expansion ;
3. Simple operation and maintenance ： In the private cloud scenario , Enterprises are both users and operators , The operation and maintenance personnel are very limited . therefore , The operation and maintenance of private cloud should be as simple as possible , At the same time, the complexity of the system will not increase greatly with the scale of the system . In this way, complex operation and maintenance can be avoided to become the bottleneck of rapid business growth , And bring hidden trouble to the stability of the system ;
4. Production ready ： When users make private cloud transformation , Usually only focus on functional features , And ignore that the private cloud is ultimately to carry the production business , In order to truly realize the enterprise data center “ Clouding ” The goal of transformation . therefore , Solutions and even supplier services “ Production ready “ Capability can guarantee the sustainability of the structural transformation 、 Go deep into , Instead of just staying in the experimental phase ;
5. The cost is reasonable ： In the end, all transformation needs to be measured in terms of economic benefits . Private cloud solutions need to focus not only on procurement costs , Also pay attention to the cost of use 、 Maintenance costs, maintenance and upgrading costs of software and hardware .

Enterprise data center transformation can only fully meet the above requirements , To be considered as a preliminary realization “ Clouding ” The goal of .

Enterprise Data Center “ Clouding ” Key points of transformation

If you want to pass IT The infrastructure upgrade fully meets the above requirements , It's not enough to use traditional architecture or just add cloud management , For example, in << China's banking information technology “ Much starker choices-and graver consequences-in ” Regulatory guidance on Development Planning >> in , The cloud transformation of bank data center puts forward the following more comprehensive requirements ：

1. Adapt to the demand of elastic change and rapid deployment of computing resources in the Internet environment ;
2. Cloud computing architecture planning , Develop cloud computing application strategy ;
3. Explore building a private cloud platform , High maturity of adoption 、 Open computing virtualization 、 Container virtualization 、 Distributed storage 、 Network virtualization and other technologies ;
4. Build a resource pool , To form a flexible supply of resources 、 Private cloud platform with flexible scheduling and dynamic metering ;

……

Combined with relevant industry reports and a large number of user practices , We also organized the following enterprise data centers “ Clouding ” The key points of transformation ：

Key points of transformation 1： From resource island to real resource pooling

In a traditional data center , The server 、 Storage and other resources for various reasons （ Isolation policy between business systems 、 Different brands and models of equipment can not be shared ）, It is difficult to share and flow resources , A single resource utilization rate tends to be polarized （ Ultra high load and no load conditions ）, In fact, the overall utilization rate of resources is relatively low .

Resource pooling is actually computing through virtualization technology 、 Storage 、 Network and other resources are virtualized , Form a unified resource pool at the data center level , Resources and data can be easily shared and flowed , At the same time, different degrees of isolation can be made through strategies . In this mode, application systems and business departments can 、 Easy access to resources , Reduce waste of resources , At the same time, it effectively improves the overall resource utilization rate .

Key points of transformation 2： From centralized to distributed architecture

In traditional data centers , Both the application layer and the infrastructure layer are centralized , The extension of centralized architecture is through scale-up （ Improve single device hardware configuration ）, In this extended mode , System performance can easily reach the bottleneck , Scalability is not good . in addition , Another problem with centralized architecture is , Rely on the reliability of a single device , The system redundancy is not enough , There is a risk of a single point of failure .

in fact , Many enterprises are considering the transformation of their business and infrastructure to a distributed architecture , Its advantage lies in not only high concurrent performance of the system , And both performance and capacity are easy to scale out , And has good fault tolerance .

Key points of transformation 3： From dedicated hardware to open software definition model

There are a lot of servers in the traditional data center 、 Storage 、 Network and other hardware equipment , There are some devices based on dedicated hardware , Like the mainframe 、 minicomputer 、 The Internet 、 Storage devices, etc , These devices are often complex 、 expensive , Slow upgrade . In addition, with the expansion of the data center , Different types of devices will greatly increase system complexity , It brings great difficulty to management .

Define products through mature software , Software and hardware can be decoupled , Free from the shackles of dedicated hardware , Unified hardware standards , Simplify management ; Software defined technology allows users to change hardware without changing hardware , Get more easily with software upgrades 、 Updated features . At the same time, the software definition pattern promotes more advanced hardware technology to be applied in the data center more quickly ; Software defined products use generic server hardware instead of dedicated hardware , The total cost of ownership of the product will also be greatly optimized .

Key points of transformation 4： From manual processing to self-service 、 Automatic service transformation

In the traditional process , In application development 、 test 、 There are many steps to go online that require developers 、 Testers 、 IT The administrator does manual processing , The whole process is complicated 、 Staggered , From development to release, the efficiency is relatively low .

The transformed business units and developers can use e-workflow , Self help application for virtual resources , The system calls the automatic deployment tools and customized templates according to the requirements , Automatic system configuration and application software installation , And related resource changes and other operations , Can effectively improve application development 、 The efficiency of resource allocation .

Key points of transformation 5： From decentralized statistics to unified measurement

Before the data center has completed resource pooling , It's difficult to have a unified view of the real-time usage of data center resources ,IT Managers have no way to make accurate analysis on the input and output of resources and make reasonable budget and planning .

Under the premise of resource pooling , With the resource metering function , It can reflect the usage of resources in real time , And accurately control resource quotas , And statistics of team resource usage ; Even according to the historical data of resource utilization, it can automatically output capacity prediction and suggestions , Help users make reasonable resource expansion planning .

The above transformation can be easily realized by using hyperfusion and cloud management software

In the last article 《 Hyper fusion is becoming the building of private clouds “ footstone ” The mainstream solution of , But do you know the relationship ？》 In the article , We've talked about the relationship between hyper fusion and private clouds . It can be seen that , Super fusion architecture with cloud management platform Can be very good to meet the enterprise data center “ Clouding ” The core needs of transformation .

The transformation of resource pool is realized by hyper fusion

1. Resource pooling is in place in one step

Hyper converged infrastructure integrates resources from multiple servers in a software defined way , For calculation 、 Storage 、 Network and other hardware resources are virtualized separately , Build a complete 、 Unified infrastructure resource pool . In short , Hyper fusion software can complete the virtualization resource pool construction of infrastructure in one time .

2. Fully distributed architecture

The core of the scheme is distributed storage , Support horizontal scaling , By adding nodes, we can get powerful concurrent processing ability , In response to changing business needs . The reliability of hyper converged infrastructure is not entirely dependent on the hardware itself , The system provides advanced fault tolerance design , Support high availability and fault self-healing function , Can effectively reduce downtime .

3. Software definition

Hyper converged infrastructure uses standard hardware （ Commercial servers and Ethernet switches ） Instead of dedicated hardware , Unbound hardware and software . meanwhile , Hardware standardization itself can greatly reduce management complexity . The software defined features enable hyperconverged architectures to iterate quickly to gain more functionality , Support the latest hardware and fully release hardware performance .

4. Integration deployment

In addition to achieving the above transformation points , Another important feature is the integrated deployment of distributed storage and virtualization , Further down, down IT The complexity of the architecture and the space occupied by the computer room , And reduce total cost of ownership .

Automation and measurable transformation by cloud management platform

Cloud management platform （Cloud Management Platform, abbreviation “CMP” ） Build in On the super fusion , Responsible for connecting developers ,IT Administrators , And unified measurement and planning of resources , among IaaS The main functions related to layer include ：

1. Automation and self service portal

CMP Provide a self-service portal , Developers can initiate self-help processes , The process can be related to the enterprise's OA perhaps ITSM Integrate , Complete resource application and approval , And by calling HCI Related interfaces , Automatically complete virtual machine creation and related configuration changes , This can avoid a lot of complicated manual operation , Greatly improve development efficiency .

2. Transparent resource management

CMP The disorderly distribution of resources can be controlled through resource quota and resource application process , And for IT Managers provide real-time resource pool usage reports and capacity planning recommendations . Through virtual machine Lifecycle Management , Timely recovery of unused resources , Reduce the waste of resources .

3. Cloud management

Private cloud and public cloud cannot replace each other , Both have unique values .“ A hybrid cloud ” It will be the enterprise of the future “ Standard configuration ”.CMP It becomes necessary to provide multi cloud management , Because it can help enterprises extend their applications from the private cloud to the edge of the public cloud , Realize application migration between public cloud and private cloud , This strategy will help companies reduce costs significantly in the future .

SmartX Project practice

SmartX In addition to the standard advantages of the above hyper fusion architecture , It also has the following unique advantages ：

1. Distributed block storage independent research and development , And experienced a lot of financial customer production environment test , It can effectively support users Complete transformation of enterprise data center ;
2. The product has good openness , Not only does it offer free native virtualization Services ELF, It can also support VMware and Citrix Virtualization platform , At the same time, in terms of server compatibility, it not only supports the mainstream commonly used by users x86 Server brand , Also can provide localization server support ;
3. SmartX Hyperfusion also has a leading edge in terms of performance and computing resource usage .

in addition ,SmartX It also provides a distributed block storage deployment mode with memory and computing separation , Flexible support of user storage resource distribution and software definition transformation .

be based on SmartX Product line and CMP To form stability 、 Simple 、 elastic 、 Open IaaS Layer scheme diagram

at present ,SmartX Has been with a number of professionals in the industry CMP Manufacturers have established ecological cooperation relationship , And complete the docking between cloud management platform and super fusion platform . After docking, the scheme further forms the following unique advantages ：

1. Stable ： Whether it's hyper fusion or CMP platform , Its core by domestic professional manufacturers and have been through many industry head customer production environment inspection , It can really support users to realize the comprehensive transformation of enterprise data center ;

2. to open up ： The whole scheme has good openness , Users can carry out hyperintegration and cloud management construction in batches according to the actual business needs , Professional cloud management manufacturers can integrate the user's original VMware System ,SmarX Hyperfusion also supports a variety of virtualization platforms and server hardware ;

3. Simple ：SmartX Hyperfusion Pools Resources in the simplest way 、 Transformation of distributed and software defined patterns ; Compared with OpenStack Complex components ,CMP The platform is easier to use .

4.. Invest on demand ： Whether hyper fusion or CMP platform , Can be built from a very small scale , Make it easier for users to initiate transformation , True realization “Think big,Start Small,Move Fast” Idea .

At present, the scheme has been applied in a large state-owned bank in China 、 A large domestic securities institution has formed a practical case , We will further introduce these user practices in the future .

SmartX For details and information about private cloud solutions, please click ：SmartX Enterprise private cloud solution