Big data as one of the most popular events at the moment, in fact, is not very new things. If we were talking about big data three or five years ago, it might have felt fresh. As one of the most important strategic resources at present, big data has been paid more and more attention by countries and enterprises. We can see it from the rise of big data to the level of national strategy. Now the knowledge sharing about big data can be said to have been overwhelming, as a beginner to query the basic information can be queried through the network. In fact, MY understanding of big data is not very rich, after all, the learning time is not very long. I’m just familiar with DKhadoop as an example to share some small knowledge to help beginners. The basic framework of big data platform is a content that many beginners must master. Because big data is too abstract, it is hard to avoid many difficulties when writing and sharing. It’s better to write through specific cases. As for the basic framework of big data platform, I still use DKhadoop, which I am familiar with, as an example. Before we do that, let’s do a simple note about DKhadoop: DKhadoop big data platform, developed by Big Fast search in order to get through the big data ecosystem and traditional non-big data companies between the channel and design of one-stop search engine level big data general computing platform (write so professional, I must be from the big fast publicity book to move over). For traditional enterprises with a large amount of data to be processed, the big data processing platform like DKhadoop can easily overcome the big data technology gap and achieve the performance of the big data platform at the search engine level. Since there are such big advantages, what is the basic framework of such a big data platform? Let’s take a look at a picture: this picture is the technical architecture of DKH standard platform
DKhadoop Big Data platform Basic framework Design Overview: 1. If you are familiar with native hadoop, you will find that dkhadoop integrates all components of the entire hadoop ecosystem. Of course, it is not only simple to integrate, but also deeply optimized and rewritten into a complete big data computing platform with higher performance. This is very different from other domestic big data platforms. DKH is the original development, while other domestic releases are just simple secondary development. 2. DKhadoop simplifies the complex configuration of big data cluster to three types of nodes (master node, management node and computing node) through middleware technology, which greatly simplifies the management operation and maintenance of cluster and enhances the high availability, high maintainability and high stability of cluster. (Data middleware is the core of DKH data exchange layer) 3. DKH is developed on the basis of original ecology, and maintains all the advantages of open source system, and is 100% compatible with open source system. In this way, big data applications developed on open source platforms can run efficiently on DKH without any changes.