Shulou(Shulou.com)06/01 Report--

Today, application platforms for large enterprises are undergoing a transformation to microservices architecture. Under the microservice architecture, the relationship between applications and underlying platforms such as databases will be reconfigured. The new generation of distributed databases must support a series of functions such as elastic scaling, resource isolation, multi-tenant, configurable consistency, multi-mode (supporting various SQL protocols), and configurable disaster recovery policies within clusters.

Capacity bottleneck of traditional single-point database is only one of the problems solved by distributed database. More importantly, under the trend of future micro-service application development and cloud platform, applications are no longer built in the "chimney" middleware plus database mode, but in a complex mesh model built by thousands or even tens of thousands of micro-service programs. Therefore, distributed databases need to meet the following capabilities in order to meet the elastic scale, high concurrency, high throughput, and flexibility requirements of upper applications.

Driven by these technical requirements, the core technical capabilities of distributed databases are divided into two aspects, one is compatibility with traditional technologies, including:

Complete ACID support, transaction and consistency assurance;

Complete support for SQL, syntax compatibility with traditional databases such as MySQL/PostgreSQL.

On the other hand, technological innovation includes:

Distributed and scalable, responding to changes in data volume and realizing elastic expansion of storage layer and computing layer;

Multi-mode access interface, supporting multi-type data management and multi-mode access interface;

HTAP transaction/analysis hybrid processing capability, under complex business requirements, to achieve physical isolation of data, mutual interference.

As a finance-level distributed relational database, the distributed database architecture of SequoiaDB giant fir database and the cloud product form oriented to microservices have helped many large financial customers including Minsheng Bank and Hengfeng Bank realize the cloud transformation and upgrading of the underlying database of a large number of business systems.

At present, the maximum number of physical nodes in a single cluster of giant fir database in the bank production system has reached 135, the maximum storage capacity of a single cluster exceeds 2.1 PB, and the maximum number of managed data in a single cluster is 131.8 billion.

SequoiaDB, as a finance-grade distributed relational database, provides a variety of important technical capabilities in the process of enterprise customer cloud architecture transformation.

Pooling of data storage resources

SequoiaDB data storage engine adopts native distributed architecture, data is completely scattered and stored among distributed nodes, automatic data distribution and management, and data can be flexibly expanded on demand.

SequoiaDB uses fragmentation technology to provide a horizontal scaling mechanism for the system, and its fragmentation process is completely transparent to applications. This mechanism solves the problem of limited hardware resources (such as memory, CPU, disk I/O) of a single server, and does not increase the complexity of application development. Giant Sequoia database can easily realize petabyte level data management through native distributed architecture. Currently, the maximum support for production environment is more than 1500 node clusters.

SequoiaDB database storage engine also implements multi-model multi-mode data management, supporting full coverage and unified management of unstructured, structured and semi-structured data. SequoiaDB's multi-mode engine design enables more diverse database platform scenarios and meets the unified management and operation and maintenance requirements for diversified business data under cloud data architecture.

At the same time, SequoiaDB provides multi-dimensional, multi-level logical and physical isolation in a large cluster. In a typical DataBase As A Service (DBaas), SequoiaDB can simultaneously serve hundreds or thousands of applications with different SLA service levels, priorities, business characteristics, and data volumes, and ensure logical and physical isolation of data between applications.

SequoiaDB provides the "Domain" feature, which can isolate the entire cluster at the physical device level, ensuring that different tables, tablespaces, and database instances are located on independent hardware devices, ensuring that high-priority online transaction applications and background statistical analysis are isolated from each other.

database instantiation

SequoiaDB supports database service instantiation.

For microservices application architectures, users can create hundreds or thousands of different relational database instances in the same cluster. The access and usage of database instances is 100% compatible with traditional relational databases, and the data used at the bottom is completely independent logically. Each instance has its own independent permission management, data management, and even can be deployed in independent hardware environments or shared devices.

SequoiaDB supports MySQL, PostgreSQL, and SparkSQL instances, as well as JSON, S3 object storage, and Posix file system instances to fully meet users 'needs for structured, semi-structured, and unstructured data.

From the perspective of application developers and DBAs, SequoiaDB provides relational database instances that are fully compatible with traditional MySQL, PostgreSQL, and SparkSQL. For example, in the MySQL instance in SequoiaDB Macrocedar database, all of its add/delete syntax, views, triggers, transactions, and even access plans are consistent with traditional MySQL.

As a distributed database, SQL instance users of SequoiaDB do not need to care whether the underlying data is spread across one or more devices. Users can simply create a partitioned table, write hundreds of millions of records to it, and its data will be automatically dispersed among different physical devices. For applications, there is no need to pay attention to sub-tables. The database automatically provides distributed transactions and distributed access.

Dual Active Disaster Recovery and Data Security

Dual-active disaster recovery refers to the technology that enables the main production database and standby database to run online at the same time in a readable and writable state in a disaster recovery system. In the bank transaction system, the dual active disaster recovery capability not only ensures that data is not lost, but also ensures that the system can be re-launched in a short time in case of an accident. Under normal circumstances, both data centers of the active-active architecture can provide read and write services for the business at the same time, and when one center goes down, all front-end applications can immediately switch to the surviving data center to continue using.

SequoiaDB has implemented a variety of disaster recovery methods at the core level, including disaster recovery strategies such as dual activity in the same city, dual centers in the same city, three centers in the same city, three centers in two places, and five centers in three places. By using the disaster recovery and high availability mechanism of SequoiaDB giant fir database, the server failure in the data center can ensure that both RTO and RPO are zero, and the entire data center or city-wide network failure can also achieve RTO=0 and RPO=0 in seconds.

About SequoiaDB Giant Sequoia Database

Giant Sequoia Database focuses on the research and development of new generation distributed database technology. Since its establishment in 2011, it has insisted on building distributed open source database engine from scratch. It is the first database vendor in China to be selected in Gartner Database Report for two consecutive years.

The main products of Giant Sequoia Database include SequoiaDB distributed relational database and SequoiaCM enterprise content management software. Enterprise application scenarios include distributed online transaction, data center, distributed content management, etc.

At present, the giant fir database has been launched in the core production business of more than 50 large commercial banks with the top 500 level, and the total number of enterprise users exceeds 1000.

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.