Skip to content

Commit

Permalink
Add v5.2 TPCH report (#6170) (#6308)
Browse files Browse the repository at this point in the history
  • Loading branch information
ti-chi-bot authored Aug 27, 2021
1 parent a1dfa08 commit 6481f58
Show file tree
Hide file tree
Showing 4 changed files with 13 additions and 13 deletions.
2 changes: 1 addition & 1 deletion TOC.md
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@
+ Benchmarks
+ [v5.1 Sysbench Performance Test Report](/benchmark/benchmark-sysbench-v5.1.0-vs-v5.0.2.md)
+ [v5.1 TPC-C Performance Test Report](/benchmark/v5.1-performance-benchmarking-with-tpcc.md)
+ [v5.1 MPP mode TPC-H 100GB Performance Test](/benchmark/v5.1-performance-benchmarking-with-tpch.md)
+ [v5.2 MPP mode TPC-H 100GB Performance Test](/benchmark/v5.2-performance-benchmarking-with-tpch.md)
+ [MySQL Compatibility](/mysql-compatibility.md)
+ [TiDB Limitations](/tidb-limitations.md)
+ [TiDB Adopters](/adopters.md)
Expand Down
Original file line number Diff line number Diff line change
@@ -1,17 +1,17 @@
---
title: TiDB TPC-H 100GB Performance Test Report -- v5.1 MPP mode vs. Greenplum 6.15.0 and Apache Spark 3.1.1
summary: Compare the TPC-H 100GB performance of TiDB 5.1 MPP mode, Greenplum 6.15.0, and Apache Spark 3.1.1.
title: TiDB TPC-H 100GB Performance Test Report -- v5.2 MPP mode vs. Greenplum 6.15.0 and Apache Spark 3.1.1
summary: Compare the TPC-H 100GB performance of TiDB 5.2 MPP mode, Greenplum 6.15.0, and Apache Spark 3.1.1.
---

# TiDB TPC-H 100GB Performance Test Report -- TiDB v5.1 MPP mode vs. Greenplum 6.15.0 and Apache Spark 3.1.1
# TiDB TPC-H 100GB Performance Test Report -- TiDB v5.2 MPP mode vs. Greenplum 6.15.0 and Apache Spark 3.1.1

## Test overview

This test aims at comparing the TPC-H 100GB performance of TiDB v5.1 in the MPP mode with that of Greenplum and Apache Spark, two mainstream analytics engines, in their latest versions. The test result shows that the performance of TiDB v5.1 in the MPP mode is two to three times faster than that of the other two solutions under TPC-H workload.
This test aims at comparing the TPC-H 100GB performance of TiDB v5.2 in the MPP mode with that of Greenplum and Apache Spark, two mainstream analytics engines, in their latest versions. The test result shows that the performance of TiDB v5.2 in the MPP mode is two to three times faster than that of the other two solutions under TPC-H workload.

In v5.0, TiDB introduces the MPP mode for [TiFlash](/tiflash/tiflash-overview.md), which significantly enhances TiDB's Hybrid Transactional and Analytical Processing (HTAP) capabilities. Test objects in this report are as follows:
In v5.0, TiDB introduces the MPP mode for [TiFlash](/tiflash/tiflash-overview.md). In v5.1 and v5.2, the MPP mode has been further improved, which significantly enhances TiDB's Hybrid Transactional and Analytical Processing (HTAP) capabilities. Test objects in this report are as follows:

+ TiDB v5.1 columnar storage in the MPP mode
+ TiDB v5.2 columnar storage in the MPP mode
+ Greenplum 6.15.0
+ Apache Spark 3.1.1 + Parquet

Expand All @@ -28,15 +28,15 @@ In v5.0, TiDB introduces the MPP mode for [TiFlash](/tiflash/tiflash-overview.md

| Service type | Software version |
|:----------|:-----------|
| TiDB | 5.1 |
| TiDB | 5.2 |
| Greenplum | 6.15.0 |
| Apache Spark | 3.1.1 |

### Parameter configuration

#### TiDB v5.1
#### TiDB v5.2

For the v5.1 cluster, TiDB uses the default parameter configuration except for the following configuration items.
For the v5.2 cluster, TiDB uses the default parameter configuration except for the following configuration items.

In the configuration file `users.toml` of TiFlash, configure `max_memory_usage` as follows:

Expand Down Expand Up @@ -87,7 +87,7 @@ The test of Apache Spark uses Apache Parquet as the storage format and stores th
>
> The following test results are the average data of three tests. All numbers are in seconds.
| Query ID | TiDB v5.1 | Greenplum 6.15.0 | Apache Spark 3.1.1 + Parquet |
| Query ID | TiDB v5.2 | Greenplum 6.15.0 | Apache Spark 3.1.1 + Parquet |
| :-------- | :----------- | :------------ | :-------------- |
| 1 | 8.08 | 64.1307 | 52.64 |
| 2 | 2.53 | 4.76612 | 11.83 |
Expand All @@ -112,11 +112,11 @@ The test of Apache Spark uses Apache Parquet as the storage format and stores th
| 21 | 24.44 | 39.08594 | 34.83 |
| 22 | 1.23 | 7.67476 | 4.59 |

![TPC-H](/media/tidb-v5-tpch-100-vs-gp-spark.png)
![TPC-H](/media/tidb-v5.2-tpch-100-vs-gp-spark.png)

In the performance diagram above:

- Blue lines represent TiDB v5.1;
- Blue lines represent TiDB v5.2;
- Red lines represent Greenplum 6.15.0;
- Yellow lines represent Apache Spark 3.1.1.
- The y-axis represents the execution time of the query. The less the time is, the better the performance is.
Binary file removed media/tidb-v5-tpch-100-vs-gp-spark.png
Binary file not shown.
Binary file added media/tidb-v5.2-tpch-100-vs-gp-spark.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.

0 comments on commit 6481f58

Please sign in to comment.