Introduction of MEAYUNDB Embedded Speed memory Database

The cloud platform is composed of n>=1 meayundb subset groups, and the applications on each subset group are identical, and the only difference is that the data stored by each MEAYUNDB subset group is different. All of your data is distributed to each subset of the cloud platform, and each subset group stores only a portion of your data.

The number of MEAYUNDB instances in the Meayundb subset group requires >=1 (the number of instances is determined by the user), and the MEAYUNDB instance data in the same subset group is identical, providing the same business applications externally, MEAYUNDB instances in the same subset group are each other's http://www.aliyun.com/zixun/aggregation/14344.html "> Data backup, which can increase the number of data backups by adding MEAYUNDB instances.

Cloud Platform does not adopt master-slave architecture, there is no single point of failure, with the expansion of the business, you can linearly increase the subset group number, improve throughput, easy to deal with hundreds of billions of line-level data storage and real-time analysis processing.

MapReduce process

After the user submits the task to the cloud platform, the cloud platform will decompose the user task, and dispatch the MEAYUNDB instance of the cloud platform, handle the user task in parallel, and finally merge the task result, and the merged result can be the input of the next round parallel computation.

MEAYUNDB mobile computing without moving data, reducing the communication overhead between client/server processes, and calculating data in memory to maximize system performance.

MEAYUNDB Performance

The software and hardware environment used in this test:

Hardware configuration: Intel (R) Xeon (r) CPU e5-2609 @ 2.40ghz,8 kernel 8 threads, memory 32GB

Operating system: Windows Server 2008 R2 Enterprise

Datasheet structure:

1. Query test:

2. Insert Test:

Value analysis of MEAYUNDB

Simple development, fast, low technical requirements, friendly to developers

High scalability, flexible expansion on demand

Seamless integration with relational databases

Reduce the impact of human factors, reduce project risk

Low latency, high concurrency, and microsecond-level data access efficiency.

Large data storage and real-time parallel computing

Management, development, maintenance cost reduction 50-80%

2-4-fold increase in productivity

10-100-fold improvement in performance