Shulou(Shulou.com)06/03 Report--

Postgres-xl is a good thing? Why?

Postgres-xl is based on postgresql database, and postgres can hang many databases.

It can do oltp, counteract mysql, mysql does not have the function of analyzing function.

It can do oltp to counteract oracle and oracle, which is ecologically weak and cannot be used in real time.

It supports json well, countering mongodb, mongodb uses fewer and fewer people.

It also terribly supports various language extensions, java, javascript, r, python, haskell.

And open source is free, simple and powerful have no friends.

Postgres-xl is the postgresql MPP cluster version, continue to hit big data database.

It is newer than the greenpulm version: it is a family with greenpulm, and it is a postgresql cluster based on MPP architecture.

But the original transformation, the version is basically the same as postgresql, the version of greenpulm can not be changed.

It is cheaper than oracle RAC/Teradata and is free to use.

It has more resources than hadoop province, no GC, high resource utilization based on C language, rich biosphere and convenient visualization.

It has an early debut and strong version stability.

Since it is so excellent, let's briefly introduce it, especially its application in big data.

Postgres-xl is divided into the following components:

A. gtm is responsible for global transactions

B. coordinator handles distribution execution

C. Datanode is responsible for the underlying processing.

Both datanode and coordinator are connected to gtm

The client connects to coordinator and runs sql

Coordinator uses gtm to distribute some transaction functions to datanode for execution

The Development Direction of big data

What is MPP?

MPP (Massively Parallel Processing), that is, massively parallel processing, is a distributed, parallel and structured database cluster with high performance, high availability and high scalability. It can provide a cost-effective general computing platform for super-large-scale data management, and is widely used to support all kinds of data warehouse systems and BI systems.

MPP architecture characteristics:

Task execution in parallel

Distributed data storage (localization)

Distributed computing

Horizontal expansion

Thinking and comparison

Characteristics of Oracle cluster: full storage per node

Features of Mysql hot backup: master-slave, full storage per node

MPP-- distributed relational database

GreenPlum: based on Postgres XL 8.2 for packaging, has not been upgraded for many years.

Mysql Cluster: just two years after coming out, the stability and performance are relatively poor.

Postgres XL and Postgres LL are two cluster modes of Postgresql database, and Postgres XL is the mainstream MPP nowadays.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.