You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@hugegraph.apache.org by ji...@apache.org on 2023/05/15 03:31:07 UTC
[incubator-hugegraph-doc] branch master updated: Update hugegraph-benchmark-0.5.6.md (#226)
This is an automated email from the ASF dual-hosted git repository.
jin pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-hugegraph-doc.git
The following commit(s) were added to refs/heads/master by this push:
new 2e5bf8c6 Update hugegraph-benchmark-0.5.6.md (#226)
2e5bf8c6 is described below
commit 2e5bf8c640d8f602c03795953c71df3164cbf4ef
Author: John Whelan <Wh...@users.noreply.github.com>
AuthorDate: Sun May 14 22:31:02 2023 -0500
Update hugegraph-benchmark-0.5.6.md (#226)
Completed conversion to English.
---
.../docs/performance/hugegraph-benchmark-0.5.6.md | 190 ++++++++++-----------
1 file changed, 93 insertions(+), 97 deletions(-)
diff --git a/content/en/docs/performance/hugegraph-benchmark-0.5.6.md b/content/en/docs/performance/hugegraph-benchmark-0.5.6.md
index bb3db47c..4df9a9e7 100644
--- a/content/en/docs/performance/hugegraph-benchmark-0.5.6.md
+++ b/content/en/docs/performance/hugegraph-benchmark-0.5.6.md
@@ -4,72 +4,67 @@ linkTitle: "HugeGraph BenchMark Performance"
weight: 1
---
-### 1 测试环境
+### 1 Test environment
-#### 1.1 硬件信息
+#### 1.1 Hardware information
| CPU | Memory | 网卡 | 磁盘 |
|----------------------------------------------|--------|-----------|-----------|
| 48 Intel(R) Xeon(R) CPU E5-2650 v4 @ 2.20GHz | 128G | 10000Mbps | 750GB SSD |
-#### 1.2 软件信息
+#### 1.2 Software information
-##### 1.2.1 测试用例
+##### 1.2.1 Test cases
-测试使用[graphdb-benchmark](https://github.com/socialsensor/graphdb-benchmarks)，一个图数据库测试集。该测试集主要包含4类测试：
+Testing is done using the [graphdb-benchmark](https://github.com/socialsensor/graphdb-benchmarks), a benchmark suite for graph databases. This benchmark suite mainly consists of four types of tests:
-- Massive Insertion，批量插入顶点和边，一定数量的顶点或边一次性提交
-- Single Insertion，单条插入，每个顶点或者每条边立即提交
-- Query，主要是图数据库的基本查询操作：
+- Massive Insertion, which involves batch insertion of vertices and edges, with a certain number of vertices or edges being submitted at once.
+- Single Insertion, which involves the immediate insertion of each vertex or edge, one at a time.
+- Query, which mainly includes the basic query operations of the graph database:
+ - Find Neighbors, which queries the neighbors of all vertices.
+ - Find Adjacent Nodes, which queries the adjacent vertices of all edges.
+ - Find Shortest Path, which queries the shortest path from the first vertex to 100 random vertices.
+- Clustering, which is a community detection algorithm based on the Louvain Method.
- - Find Neighbors，查询所有顶点的邻居
- - Find Adjacent Nodes，查询所有边的邻接顶点
- - Find Shortest Path，查询第一个顶点到100个随机顶点的最短路径
+##### 1.2.2 Test dataset
-- Clustering，基于Louvain Method的社区发现算法
+Tests are conducted using both synthetic and real data.
-##### 1.2.2 测试数据集
-
-测试使用人造数据和真实数据
-
-- MIW、SIW和QW使用SNAP数据集
+- MIW, SIW, and QW use SNAP datasets:
- [Enron Dataset](http://snap.stanford.edu/data/email-Enron.html)
-
- [Amazon dataset](http://snap.stanford.edu/data/amazon0601.html)
-
- [Youtube dataset](http://snap.stanford.edu/data/com-Youtube.html)
-
- [LiveJournal dataset](http://snap.stanford.edu/data/com-LiveJournal.html)
-- CW使用[LFR-Benchmark generator](https://sites.google.com/site/andrealancichinetti/files)生成的人造数据
+- CW uses synthetic data generated by the [LFR-Benchmark generator](https://sites.google.com/site/andrealancichinetti/files).
-###### 本测试用到的数据集规模
+The size of the datasets used in this test are not mentioned.
-| 名称 | vertex数目 | edge数目 | 文件大小 |
+| Name | Number of Vertices | Number of Edges | File Size |
|-------------------------|-----------|-----------|--------|
| email-enron.txt | 36,691 | 367,661 | 4MB |
| com-youtube.ungraph.txt | 1,157,806 | 2,987,624 | 38.7MB |
| amazon0601.txt | 403,393 | 3,387,388 | 47.9MB |
| com-lj.ungraph.txt | 3997961 | 34681189 | 479MB |
-#### 1.3 服务配置
+#### 1.3 Service configuration
-- HugeGraph版本：0.5.6，RestServer和Gremlin Server和backends都在同一台服务器上
+- HugeGraph version: 0.5.6, RestServer and Gremlin Server and backends are on the same server
- - RocksDB版本：rocksdbjni-5.8.6
+ - RocksDB version: rocksdbjni-5.8.6
-- Titan版本：0.5.4, 使用thrift+Cassandra模式
+- Titan version: 0.5.4, using thrift+Cassandra mode
- - Cassandra版本：cassandra-3.10，commit-log 和 data 共用SSD
+ - Cassandra version: cassandra-3.10, commit-log and data use SSD together
-- Neo4j版本：2.0.1
+- Neo4j version: 2.0.1
-> graphdb-benchmark适配的Titan版本为0.5.4
+> The Titan version adapted by graphdb-benchmark is 0.5.4.
-### 2 测试结果
+### 2 Test results
-#### 2.1 Batch插入性能
+#### 2.1 Batch insertion performance
| Backend | email-enron(30w) | amazon0601(300w) | com-youtube.ungraph(300w) | com-lj.ungraph(3000w) |
|-----------|------------------|------------------|---------------------------|-----------------------|
@@ -77,24 +72,24 @@ weight: 1
| Titan | 10.15 | 108.569 | 150.266 | 1217.944 |
| Neo4j | 3.884 | 18.938 | 24.890 | 281.537 |
-_说明_
+_Instructions_
-- 表头"（）"中数据是数据规模，以边为单位
-- 表中数据是批量插入的时间，单位是s
-- 例如，HugeGraph使用RocksDB插入amazon0601数据集的300w条边，花费5.711s
+- The data scale is in the table header in terms of edges
+- The data in the table is the time for batch insertion, in seconds
+- For example, HugeGraph(RocksDB) spent 5.711 seconds to insert 3 million edges of the amazon0601 dataset.
-##### 结论
+##### Conclusion
-- 批量插入性能 HugeGraph(RocksDB) > Neo4j > Titan(thrift+Cassandra)
+- The performance of batch insertion: HugeGraph(RocksDB) > Neo4j > Titan(thrift+Cassandra)
-#### 2.2 遍历性能
+#### 2.2 Traversal performance
-##### 2.2.1 术语说明
+##### 2.2.1 Explanation of terms
-- FN(Find Neighbor), 遍历所有vertex, 根据vertex查邻接edge, 通过edge和vertex查other vertex
-- FA(Find Adjacent), 遍历所有edge，根据edge获得source vertex和target vertex
+- FN(Find Neighbor): Traverse all vertices, find the adjacent edges based on each vertex, and use the edges and vertices to find the other vertices adjacent to the original vertex.
+- FA(Find Adjacent): Traverse all edges, get the source vertex and target vertex based on each edge.
-##### 2.2.2 FN性能
+##### 2.2.2 FN performance
| Backend | email-enron(3.6w) | amazon0601(40w) | com-youtube.ungraph(120w) | com-lj.ungraph(400w) |
|-----------|-------------------|-----------------|---------------------------|----------------------|
@@ -102,11 +97,11 @@ _说明_
| Titan | 8.084 | 92.507 | 184.543 | 1099.371 |
| Neo4j | 2.424 | 10.537 | 11.609 | 106.919 |
-_说明_
+_Instructions_
-- 表头"（）"中数据是数据规模，以顶点为单位
-- 表中数据是遍历顶点花费的时间，单位是s
-- 例如，HugeGraph使用RocksDB后端遍历amazon0601的所有顶点，并查找邻接边和另一顶点，总共耗时45.118s
+- The data in the table header "( )" represents the data scale, in terms of vertices.
+- The data in the table represents the time spent traversing vertices, in seconds.
+- For example, HugeGraph uses the RocksDB backend to traverse all vertices in amazon0601, and search for adjacent edges and another vertex, which takes a total of 45.118 seconds.
##### 2.2.3 FA性能
@@ -116,25 +111,25 @@ _说明_
| Titan | 7.361 | 93.344 | 169.218 | 1085.235 |
| Neo4j | 1.673 | 4.775 | 4.284 | 40.507 |
-_说明_
-
-- 表头"（）"中数据是数据规模，以边为单位
-- 表中数据是遍历边花费的时间，单位是s
-- 例如，HugeGraph使用RocksDB后端遍历amazon0601的所有边，并查询每条边的两个顶点，总共耗时10.764s
+_Explanation_
-###### 结论
+- The data size in the header "( )" is based on the number of vertices.
+- The data in the table is the time it takes to traverse the vertices, in seconds.
+- For example, HugeGraph with RocksDB backend traverses all vertices in the amazon0601 dataset, and looks up adjacent edges and other vertices, taking a total of 45.118 seconds.
+-
+###### Conclusion
-- 遍历性能 Neo4j > HugeGraph(RocksDB) > Titan(thrift+Cassandra)
+- Traversal performance: Neo4j > HugeGraph(RocksDB) > Titan(thrift+Cassandra)
-#### 2.3 HugeGraph-图常用分析方法性能
+#### 2.3 Performance of Common Graph Analysis Methods in HugeGraph
-##### 术语说明
+##### Terminology Explanation
-- FS(Find Shortest Path), 寻找最短路径
-- K-neighbor，从起始vertex出发，通过K跳边能够到达的所有顶点, 包括1, 2, 3...(K-1), K跳边可达vertex
-- K-out, 从起始vertex出发，恰好经过K跳out边能够到达的顶点
+- FS (Find Shortest Path): finding the shortest path between two vertices
+- K-neighbor: all vertices that can be reached by traversing K hops (including 1, 2, 3...(K-1) hops) from the starting vertex
+- K-out: all vertices that can be reached by traversing exactly K out-edges from the starting vertex.
-##### FS性能
+##### FS performance
| Backend | email-enron(30w) | amazon0601(300w) | com-youtube.ungraph(300w) | com-lj.ungraph(3000w) |
|-----------|------------------|------------------|---------------------------|-----------------------|
@@ -142,64 +137,65 @@ _说明_
| Titan | 11.818 | 0.239 | 377.709 | 575.678 |
| Neo4j | 1.719 | 1.800 | 1.956 | 8.530 |
-_说明_
+_Explanation_
-- 表头"（）"中数据是数据规模，以边为单位
-- 表中数据是找到**从第一个顶点出发到达随机选择的100个顶点的最短路径**的时间，单位是s
-- 例如，HugeGraph使用RocksDB后端在图amazon0601中查找第一个顶点到100个随机顶点的最短路径，总共耗时0.103s
+- The data in the header "()" represents the data scale in terms of edges
+- The data in the table is the time it takes to find the shortest path **from the first vertex to 100 randomly selected vertices** in seconds
+- For example, HugeGraph using the RocksDB backend to find the shortest path from the first vertex to 100 randomly selected vertices in the amazon0601 graph took a total of 0.103s.
-###### 结论
+###### Conclusion
-- 在数据规模小或者顶点关联关系少的场景下，HugeGraph性能优于Neo4j和Titan
-- 随着数据规模增大且顶点的关联度增高，HugeGraph与Neo4j性能趋近，都远高于Titan
+- In scenarios with small data size or few vertex relationships, HugeGraph outperforms Neo4j and Titan.
+- As the data size increases and the degree of vertex association increases, the performance of HugeGraph and Neo4j tends to be similar, both far exceeding Titan.
-##### K-neighbor性能
+##### K-neighbor Performance
-顶点 | 深度 | 一度 | 二度 | 三度 | 四度 | 五度 | 六度
------ | -- | ------ | ------ | ------ | ------ | ------ | ---
-v1 | 时间 | 0.031s | 0.033s | 0.048s | 0.500s | 11.27s | OOM
-v111 | 时间 | 0.027s | 0.034s | 0.115 | 1.36s | OOM | --
-v1111 | 时间 | 0.039s | 0.027s | 0.052s | 0.511s | 10.96s | OOM
+Vertex | Depth | Degree 1 | Degree 2 | Degree 3 | Degree 4 | Degree 5 | Degree 6
+----- | ----- | -------- | -------- | -------- | -------- | -------- | --------
+v1 | Time | 0.031s | 0.033s | 0.048s | 0.500s | 11.27s | OOM
+v111 | Time | 0.027s | 0.034s | 0.115s | 1.36s | OOM | --
+v1111 | Time | 0.039s | 0.027s | 0.052s | 0.511s | 10.96s | OOM
-_说明_
+_Explanation_
-- HugeGraph-Server的JVM内存设置为32GB，数据量过大时会出现OOM
+- HugeGraph-Server's JVM memory is set to 32GB and may experience OOM when the data is too large.
-##### K-out性能
+##### K-out performance
-顶点 | 深度 | 一度 | 二度 | 三度 | 四度 | 五度 | 六度
+Vertex | Depth | 1st Degree | 2nd Degree | 3rd Degree | 4th Degree | 5th Degree | 6th Degree
----- | -- | ------ | ------ | ------ | ------ | --------- | ---
-v1 | 时间 | 0.054s | 0.057s | 0.109s | 0.526s | 3.77s | OOM
- | 度 | 10 | 133 | 2453 | 50,830 | 1,128,688 |
-v111 | 时间 | 0.032s | 0.042s | 0.136s | 1.25s | 20.62s | OOM
- | 度 | 10 | 211 | 4944 | 113150 | 2,629,970 |
-v1111 | 时间 | 0.039s | 0.045s | 0.053s | 1.10s | 2.92s | OOM
- | 度 | 10 | 140 | 2555 | 50825 | 1,070,230 |
+v1 | Time | 0.054s | 0.057s | 0.109s | 0.526s | 3.77s | OOM
+ | Degree | 10 | 133 | 2453 | 50,830 | 1,128,688 |
+v111 | Time | 0.032s | 0.042s | 0.136s | 1.25s | 20.62s | OOM
+ | Degree | 10 | 211 | 4944 | 113150 | 2,629,970 |
+v1111 | Time | 0.039s | 0.045s | 0.053s | 1.10s | 2.92s | OOM
+ | Degree | 10 | 140 | 2555 | 50825 | 1,070,230 |
+
-_说明_
+_Explanation_
-- HugeGraph-Server的JVM内存设置为32GB，数据量过大时会出现OOM
+- The JVM memory of HugeGraph-Server is set to 32GB, and OOM may occur when the data is too large.
-###### 结论
+###### Conclusion
-- FS场景，HugeGraph性能优于Neo4j和Titan
-- K-neighbor和K-out场景，HugeGraph能够实现在5度范围内秒级返回结果
+- In the FS scenario, HugeGraph outperforms Neo4j and Titan in terms of performance.
+- In the K-neighbor and K-out scenarios, HugeGraph can achieve results returned within seconds within 5 degrees.
-#### 2.4 图综合性能测试-CW
+#### 2.4 Comprehensive Performance Test - CW
-| 数据库 | 规模1000 | 规模5000 | 规模10000 | 规模20000 |
+| Database | Size 1000 | Size 5000 | Size 10000 | Size 20000 |
|-----------------|--------|---------|----------|----------|
| HugeGraph(core) | 20.804 | 242.099 | 744.780 | 1700.547 |
| Titan | 45.790 | 820.633 | 2652.235 | 9568.623 |
| Neo4j | 5.913 | 50.267 | 142.354 | 460.880 |
-_说明_
+_Explanation_
-- "规模"以顶点为单位
-- 表中数据是社区发现完成需要的时间，单位是s，例如HugeGraph使用RocksDB后端在规模10000的数据集，社区聚合不再变化，需要耗时744.780s
-- CW测试是CRUD的综合评估
-- 该测试中HugeGraph跟Titan一样，没有通过client，直接对core操作
+- The "scale" is based on the number of vertices.
+- The data in the table is the time required to complete community discovery, in seconds. For example, if HugeGraph uses the RocksDB backend and operates on a dataset of 10,000 vertices, and the community aggregation is no longer changing, it takes 744.780 seconds.
+- The CW test is a comprehensive evaluation of CRUD operations.
+- In this test, HugeGraph, like Titan, did not use the client and directly operated on the core.
-##### 结论
+##### Conclusion
-- 社区聚类算法性能 Neo4j > HugeGraph > Titan
+- Performance of community detection algorithm: Neo4j > HugeGraph > Titan