Overview

DataArts Studio DataArts DataService aims to build a unified data service bus for enterprises to centrally manage internal and external API services. DataArts DataService helps you quickly generate data APIs based on data tables and allows you manage the full lifecycle of APIs, covering API publishing, management, and O&M. With DataArts DataService, you can implement microservice aggregation, frontend-backend separation, and system integration, and provide functions and data for partners and developers easily and quickly at a low cost and risk.

DataArts DataService has the following advantages over other data sharing and exchange methods:

Unified interface standards reduce the workload for interconnection with upper-layer applications.
Data logic is deployed on the data platform and is therefore decoupled from the application logic. This reduces repeated development of data models and avoids frequent changes caused by data logic adjustment.
Data logic-related storage and compute resources are deployed on the data platform, reducing resource consumption on applications.
A large amount of detailed and sensitive data is inaccessible to applications. In addition, DataArts DataService improves data security by means of API review and publishing, authentication and throttling, and dynamic anonymization.

DataArts DataService encapsulates data logic into RESTful APIs of a unified standard that can be used to access data. DataArts DataService applies to quick response to the requests for accessing a small amount of data. To open a large amount of data, you are advised to adopt data sharing and exchange or other solutions.

Publishing an API

To publish an API or a group of APIs, do as follows:

Buying and Managing an Exclusive Cluster
If you want to use DataArts DataService, you must buy a DataArts DataService Exclusive cluster.
Creating a Reviewer in DataArts DataService
Before creating an API, you need to create a reviewer.
Creating an API
You can generate an API. An API can be generated using configuration or a script/MyBatis.
Debugging an API
Debug the created API on the management console to check whether it runs properly.
Publishing an API
The API can be called only after it is published.
Managing APIs
You can manage the published API as needed.
Orchestrating an API
API orchestration allows you to reorganize and reconstruct APIs in a visualized manner based on specific service logic and processes without compiling code. In this way, you can perform secondary development easily without affecting native APIs.
(Optional) Configuring a Throttling Policy
To ensure the stability of backend services, you can perform throttling on the API.
(Optional) Authorizing an API
An app defines the identity of an API caller. An API that uses app or IAM authentication must be authorized so that the authentication information for calling the API can be obtained.

Calling an API

To call an API, perform the following operations:

Obtain an API.
Obtain the API from the service catalog. An API can be called only after it is published.
Applying for API Authorization
If you are an API developer and want to call an API which uses app or IAM authentication, you must apply for API authorization.
Calling the API.
After completing the preceding steps, you can call the API.

Overview Page

On the Overview page, you can view various monitoring data views. The Overview page displays information about APIs and apps.

Figure 1 API-based data statistics
Click to enlarge

**Table 1** API-based data statistics
Category	Metric	Description
Data overview	Published	Number of APIs that have been published
	Developing	Number of APIs that are being developed
	Applications	Number of apps authorized by published APIs
	Total calls	Number of calls to the APIs in all clusters over the last seven days (excluding the current day)
	Successful	Number of successful API calls
	Failed	Number of failed API calls
	Valid	Number of valid API calls. Valid API calls refer to the calls that pass the verification.
	Invalid	Number of invalid API calls. Invalid API calls refer to the calls that fail to pass the verification due to incorrect request parameters.
Trend chart	APIs Called	Trend of the number of API calls in one or all clusters during a specified time period. Time period: last 12 hours, last 1 day, last 7 days, or last 30 days Cluster dimension: one cluster or all clusters Number of calls: total number of calls, number of successful/failed calls, or number of valid/invalid calls
Trend chart	APIs Published	Trend of the number of API publishing times in a specified time period Time period: today, this week, this month, or this year
Top 5 statistics	Top 5 APIs by Call Rate	Top 5 APIs by call rate in one or all clusters during a specified time period Time period: last 12 hours, last 1 day, last 7 days, or last 30 days Cluster dimension: one cluster or all clusters Rate: success rate, failure rate, validity rate, or invalidity rate
	Top 5 APIs by Call Duration	Top 5 APIs by call duration in one or all clusters during a specified time period Time period: last 12 hours, last 1 day, last 7 days, or last 30 days Cluster dimension: one cluster or all clusters Duration: average duration, average duration of successful calls, or average duration of failed calls
	Top 5 APIs by Call Quantity	Top 5 APIs by the number of calls in one or all clusters during a specified time period. (If an API is authorized to different apps, the number of calls made by those apps to the API is counted together.) Time period: last 12 hours, last 1 day, last 7 days, or last 30 days Cluster dimension: one cluster or all clusters Number of calls: total number of calls, number of successful/failed calls, or number of valid/invalid calls
	Top 5 by Call Quantity	Top 5 APIs by the number of calls in one or all clusters during a specified time period. (If an API is authorized to different apps, the number of calls made by those apps to the API is counted separately.) Time period: last 12 hours, last 1 day, last 7 days, or last 30 days Cluster dimension: one cluster or all clusters Number of calls: total number of calls, number of successful/failed calls, or number of valid/invalid calls

Figure 2 App-based data statistics
Click to enlarge

**Table 2** App-based data statistics
Category	Metric	Description
Data overview	Completed	Number of apps authorized by all APIs
	Total calls	Number of calls to the APIs (that use app or IAM authentication) in all clusters over the last seven days (excluding the current day)
	Successful	Number of successful calls to APIs that use app or IAM authentication.
	Failed	Number of failed calls to APIs that use app or IAM authentication.
	Valid	Number of valid calls to APIs that use app or IAM authentication. Valid API calls refer to the calls that pass the verification.
	Invalid	Number of invalid calls to APIs that use app or IAM authentication. Invalid API calls refer to the calls that fail to pass the verification due to incorrect request parameters.
Trend chart	APIs Called	Trend of the number of calls to APIs that use app or IAM authentication in all clusters during a specified time period. Time period: last 12 hours, last 1 day, last 7 days, or last 30 days Number of calls: total number of calls, number of successful/failed calls, or number of valid/invalid calls
Top 5 statistics	Top 5 Apps by Call Rate	Top 5 APIs (using app or IAM authentication) by call rate in all clusters during a specified time period Time period: last 12 hours, last 1 day, last 7 days, or last 30 days Rate: success rate, failure rate, validity rate, or invalidity rate
	Top 5 Apps by Call Duration	Top 5 APIs (using app or IAM authentication) by call duration in all clusters during a specified time period Time period: last 12 hours, last 1 day, last 7 days, or last 30 days Duration: average duration, average duration of successful calls, or average duration of failed calls
	Top 5 Apps by Call Quantity	Top 5 APIs (using app or IAM authentication) by the number of calls in all clusters during a specified time period. (If an API is authorized to different apps, the number of calls made by those apps to the API is counted together.) Time period: last 12 hours, last 1 day, last 7 days, or last 30 days Number of calls: total number of calls, number of successful/failed calls, or number of valid/invalid calls
	Top 5 by Call Quantity	Top 5 APIs (using app or IAM authentication) by the number of calls in all clusters during a specified time period. (If an API is authorized to different apps, the number of calls made by those apps to the API is counted separately.) Time period: last 12 hours, last 1 day, last 7 days, or last 30 days Number of calls: total number of calls, number of successful/failed calls, or number of valid/invalid calls