Software deployment overview

1. Server configuration recommendations

Mapmost Studio theoretically supports deployment in all mainstream x86 architecture Linux operating system distributions. However, taking into account the stability of the distribution, support cycle, community activity and other factors, in the absence of other specific requirements (such as the requirements of domestic Xinchuang), it is recommended to use Ubuntu 22.04/24.04 to deploy the system.

Operating System	Hardware
Ubuntu 22.04/24.04	CPU：Intel(R) Xeon(R) CPU E5-2650 v4
	Memory: 32 GB
	Disk: SSD 1T
	Network: Network 1000baseT/Full

2. System service list

• Business Application:

application	Default port	illustrate
mapmost-studio-web	/	Studio front end
mapmost-studio-carto	/	Drawing platform front end
mapmost-studio-vector-map	8222	Vector map service
cloud-minio-upload	18092	File upload service
modeldeployment-manager	8101, 5555 (after v9.1.0)	Slice release management service
studio-core-master	8100	Studio core management service (new in v9.1.0 version)

• middleware:

middleware	Default port	illustrate	Current version in use
Docker	/	Container Engine	24.0.9
OpenResty	80、8085	Web server, Lua script	1.21.4.1-1
MySQL	3306	Database	8.0.33
Redis	6379	Cache	6.2.12
PostgreSQL（PostGIS）	5432	Spatial-temporal database	12~17
MinIO	9000	OBJect Storage	2021-06-17T00-10-46Z

• System Services:

Serve	illustrate
NFS	File sharing (new in v9.1.0, required for multi-compute node deployment)

3. System deployment architecture

Currently, the Mapmost Studio system supports three deployment architectures, which can be selected according to specific circumstances:

(1) Single node deployment

All services are deployed on the same server.

It is easy to manage, but requires higher server hardware configuration, and the impact will be greater when the server fails.

(2) Multi-node deployment (single computing node)

Deploy front-end and back-end application services and middleware services separately on different servers.

Reduce resource competition on the server, and flexibly allocate hardware resources to the server according to the purpose. The impact of failure is small, but the cost and management complexity are high.

An example of server node allocation is as follows:

node	use	Deployment service
node1	Application Service Node	Docker, OpenResty, front-end and back-end services
node2	Database node	Docker、MySQL、Redis、PostgreSQL
node3	File node	Docker、MinIO

Tips：

Each middleware service can also be independently deployed on multiple server nodes to further reduce the impact of server failure.

(3) Multi-node deployment (multiple computing nodes)

Based on the previous deployment structure, the application service node is split into management nodes and computing nodes. The computing nodes can be horizontally expanded as needed, thereby improving the overall performance of the system.

An example of server node allocation is as follows:

node	use	Deployment service
node1	Management Node	Docker, OpenResty, front-end and back-end services
node2	Compute node	Docker, modeldeployment-manager service
node3	Compute node	Docker, modeldeployment-manager service
node4	Database node	Docker、MySQL、Redis、PostgreSQL
node5	File node	Docker, MinIO, NFS services

---