Productization, standardization, and industrialization are relatively heavy topics, and it is also a pain point for our domestic software to go abroad and enter the world. I did not participate in the business negotiation, nor did I participate in the selection of the entire data warehouse at that time. This may be my regret. It is said that when the selection was made at that time, the customer directly asked to choose from IBM and Teradata. As we all know, These two major manufacturers are the two best data warehouse solutions in the world, and they are also the two best manufacturers of productization, standardization and industrialization. This is not to blame them. They provide an integrated solution for the construction of the entire data warehouse. The solution, including scheduling, ETL, all-in-one, DWM models, reports, and mining, is supported by a series of tools and methodologies, but the price is also expensive. Due to the project, I can only find a similar diagram from the Internet to illustrate the integrated solution of IBM and Teradata.
IBM all-in-one solution, not all tools are shown in this diagram
Teradata Warehouse Framework
You can see that in the construction of the entire data warehouse, these two manufacturers have done a very good job, and this is what we should learn from.
Then I will talk about the current situation in China. I have never participated in domestic BI projects, so I cannot make a unified comparison. I only explain what I have learned.
(1) Our department should have done several overseas BI projects, but when I came to our department, the entire department did not have an integrated solution for BI projects, nor did it generate its own products and methodology, to be honest In other words, the department seems to be stuck in a state of doing things for the sake of doing things.
This overseas project is a drawback. We all know that European customers are not only concerned about what you can do, but also why you do it and what is the reason for you to do it. The latter two are their comparisons. Attention, which we lacked for 3 years ago.
(2) Products, what everyone must understand is what is a product, a product is standardized and industrialized, not dependent on the project, and the project should be more in line with your product, rather than making the product depend on your project, Each project can have personalized things, but the product should cover at least 80% of the entire needs, but three years ago, our team, I think there was no product, but just a tool.
(3) Methodology (model), this is relatively heavy, everyone knows that data warehouses have models, but everyone does it only for needs, not systematic and structured, I remember that when I first participated in the model, it was not because of me. I understand the theory of models, but because I understand the business, the whole modeling theory is only known when I read a book that a client gave my colleague. At this time, I realized the reason why IBM's model was not imagined out of thin air. He has extensive theories as support, and can quickly resonate with customers, allowing customers to understand why they need such a model, even a layman can understand the basis of the entire model.
(4) Integrated solution, the previous blockbuster mentioned the integrated solution, the data warehouse has a characteristic, that is, the dependence of product tools is relatively serious, because we do not have our own precipitation, so there is a phenomenon called taking This leads to the high cost of implementation, and the manpower we invest is not proportional to the income, and the three-party manufacturers make a lot of money.
Summarize:
(1) The data warehouse solutions of traditional application industries (telecommunications, banks) have now been achieved by foreign manufacturers to the extreme. Foreign customers attach importance to stability and security, and customers will give priority to purchasing overseas mature solutions and systems. We can only implement role participation
(2) The domestic software industry must pay attention to the combination of software and related knowledge fields, so that it can resonate with customers when explaining it.
(3) You must have your own methodology, even if you learn from others, you must incorporate the methodology into your product
Vision
Now the data warehouse based on big data solutions, domestic and foreign manufacturers are now on the same starting line, and I hope to see the emergence of domestic manufacturers. .