一个神奇的ETL工具：CR-X

  
   虽然， 我的SWING和规则引擎技术还没用研究清楚， 但是最近接受到总部的一个任务，要求写一个数据模型规范， 真是一个非常困难的任务， 因为我基本上没用任何数据库设计的背景， 而且我的数学学习的也不好， 但是老板已经接下这个任务，那么我就必须得突破这个难关。好在老板同意申请专项资金，请第三方公司咨询完成。

    
    这个星期就向所有前期联系的IT公司发邮件联系，不过大家一听是软课题， 建立数据模型，而并非开发系统，很多公司都打了退堂鼓，不过有一家神奇的印度公司却表现出很大的热情， 这家公司在神奇的Bangalore  中文译名：班加罗尔。 在交流中，他们推荐了一款神奇的软件CR-X，用于实现ETL的功能，我问了很多人，都不知道这个软件，后来没办法，找了一个HUST的教授，才了解到这个软件， 地址是
    CR-X是历史比较悠久的ETL中间件了。具体网址 http://www.cr-x.com/
具体类似开源的中间件得找一下，一下没关注这块。
    数据仓库本身我们项目组倒是很少研究，但是ETL，bigdata、流数据和数据
分析倒是在做一些工作。

     于是立马去看了一下，感觉这个组件还是非常实用的，并且他的公司在美丽的Melbourne。 下面是一个简单的介绍，后续还会补充一些收集的资料。



Solution Design & Architecture
Just as building architectures are best envisioned in terms of a number of complementary views or models, so too are software architectures. In particular, structural views help document and communicate the architecture in terms of the components and their relationships, and are useful in assessing architectural qualities like extensibility.

Because CR-X configurations are so rapid to develop, handle masses of data and typically interact with dozens of upstream and downstream sources and systems, a well-defined architecture is a key element in a successful CR-X installation.





hi My-friend :
            I am a product manager in CMH, which is a local branch of the bigest wireless operator in china.
           we are going to rebuild our wireless network analysis and Optimization system, because our existing system has greeat  bottle-neck in data-base componment.  First of all,  we have to wait a long time to query dauly network prfoemance data; second , the data storage duration is 42days only, yet sometimes we have to fetch some network performance data 1 years before.  Lastly,  third party information system could not share the data, as a result ,we have to develop and maintain a lot of interfaces .  
         in order to solve these three problems, we have to  redesign the data-base management systm first, in other words, design the data model first.  we hope the data model is easy to understand , easy to change, efficiency to query, and easy to share.  as we can see, in such data analysis system, data model is so important, therefor we decide to select
Professional consulting firm to do this project.  so may I know CR-X is appropriate to do such project. we wish to cousult some problem first.
1) in CR-X, what is the general method , process , and tools to build data model.
2) how can we  evaluate the quality of a data model
3) how can we verify the Prototype, for example via a data-base product.
4) how can we manage the artifact of the data model, because, the data model will be change as network scale extends.
5) we have about 100 thousand Gsm Cell and 3 thousand 3G cell. for each cell, we will record about 200 performce data with hour interval. so inorder to build a data model for such system, what will be the average budget?
      by the way, we are interesting to know whether you have developed some infomation systems for the operator, for example the Telstra,  OPTUS,or the 3 Hutchison . if you have such experience,would you  please let  me see some of the Prototype?  I just want to know the compentence of CR-X form such prototype.
      please keep in touch and hope to get your reply.

flyqantas