Update doc

Signed-off-by: Sijie Shen <[email protected]>

README.md

@@ -1,8 +1,8 @@
 # GART: Graph Analysis on Relational Transactional Datasets
 
-GART is a graph extension that includes an interface to an RDBMS and a dynamic graph store for online graph processing. It is designed to bridge the gap between relational OLTP and graph-based OLAP.
+GART is a graph extension that includes an interface to an RDBMS and a dynamic graph store for online graph computation. It is designed to bridge the gap between relational OLTP and graph-based OLAP.
 
-Please to refer [GART documentation](https://graphscope.github.io/GART/documentation/getting-started/quick-start.html) for more details.
+Please to refer [GART documentation](https://graphscope.github.io/GART) for more details.
 
 ## Table of Contents
 - [What is GART](#what-is-gart)
@@ -48,7 +48,7 @@ GART captures data changes from data sources by logs (e.g., Binlogs in SQL syste
   This sample records the log that inserts a tuple of `organisation`.
 
 - **2. Model Convert (RGMapping Converter)**:
-This step is an important step for GART. The conversion between different data models for HTGAP workloads requires more semantic information.
+This step is an important step for GART. The conversion between different data models for online graph computation requires more semantic information.
 For example, it needs the mapping between relational tables and vertex/edge types, and the mapping between relational attributes and vertex/edge properties.
 The GART administrator (such as DBA) can define the rules of relation-graph mapping (RGMapping) once by the interfaces provided by GART.
 GART will convert relational data changes into graph data changes in the *unified logs* (UnifiedLog) automatically.
@@ -66,7 +66,7 @@ To adapt to rich workload flexibility, GART proposes transparent data model conv
 We provide a sample definition file called [rgmapping-ldbc.yaml](vegito/test/schema/rgmapping-ldbc.yaml).
 
 ### Efficient Dynamic Graph Storage
-To ensure the performance of graph analytical processing (GAP), GART proposes an efficient dynamic graph storage with good locality that stems from key insights into HTGAP workloads, including:
+To ensure the performance of graph analytical processing (GAP), GART proposes an efficient dynamic graph storage with good locality that stems from key insights into online graph computation, including:
 1. an efficient and mutable compressed sparse row (CSR) representation to guarantee the locality of scanning edges;
 2. a coarse-grained MVCC to reduce the temporal and spatial overhead of versioning;
 3. a flexible property storage to efficiently run various GAP workloads.

docs/docs.rst

@@ -4,7 +4,7 @@
    contain the root `toctree` directive.
 
 .. meta::
-   :description: GART is a graph extension that includes an interface to an RDBMS and a dynamic graph store for online graph processing.
+   :description: GART is a graph extension that includes an interface to an RDBMS and a dynamic graph store for online graph computation.
    :keywords: distributed-systems, distributed, graph-analytics, in-memory-storage, big-data-analytics, distributed-comp
 
 GART: Graph Analysis on Relational Transactional Datasets

docs/documentation/getting-started/intro-content.rst

@@ -14,13 +14,13 @@ There are two common architectures for graph data processing:
 
    Common architectures for graph data processing
 
-To do graph computation, relational database systems will provide some interface extensions, which are equivalent to a kind of syntactic sugar for graph processing, and the underlying still uses the relational model, e.g., `GraphScript`_ provided by SAP HANA, `SQL Graph`_ provided by SQL Server, `Apache AGE`_ as an extension for PostgreSQL.
+To do graph computation, relational database systems will provide some interface extensions, which are equivalent to a kind of syntactic sugar for graph computation, and the underlying still uses the relational model, e.g., `GraphScript`_ provided by SAP HANA, `SQL Graph`_ provided by SQL Server, `Apache AGE`_ as an extension for PostgreSQL.
 
-On the other hand, there is also the option of exporting the data as offline graph data for offline graph data analysis and processing on a graph processing system (`Neo4j`_, `GraphScope`_, ...).
+On the other hand, there is also the option of exporting the data as offline graph data for offline graph data analysis and processing on a graph computation system (`Neo4j`_, `GraphScope`_, ...).
 
 An ideal model for processing online graph data needs to fulfill at least the following three requirements:
 
-1. **Performance**. Storing data as tables in relational storage for graph processing requires a large number of time-consuming join operations. Such operations cause serious performance degradation compared to the graph native storage and execution engine.
+1. **Performance**. Storing data as tables in relational storage for graph computation requires a large number of time-consuming join operations. Such operations cause serious performance degradation compared to the graph native storage and execution engine.
 
 2. **Data freshness**. To keep the data consistent, long offline data moves need to be avoided so that analysis can be performed quickly on the updated data. In some time-sensitive scenarios, such as fraud detection, early analysis results can avoid losses.
 
@@ -29,7 +29,7 @@ An ideal model for processing online graph data needs to fulfill at least the fo
 What is GART?
 -----------------
 
-GART is a graph extension that includes an interface to an RDBMS and a dynamic graph store for online graph processing.
+GART is a graph extension that includes an interface to an RDBMS and a dynamic graph store for online graph computation.
 
 There is a need to utilize graph data flexibly without altering the existing relational database system. Additionally, the issue of transparency arises, where users should not need to be concerned with the intricacies of graph data storage or the process of maintaining data consistency between relational and graph data stores.