Benchmark

On this page, different database backends of Gorse are compared.

RESTful APIs

Before benchmarking RESTful APIs, the develop environment should be set up first. Then, change the working directory to server and run the benchmark script:

cd server
bash bench_test.sh --cache redis --data mysql

• The cache argument should be one of redis, mysql, postgres or mongodb.
• The data argument should be one of clickhouse, mysql, postgres or mongodb.

The benchmark result (in milliseconds) is as follows:

Database[1]
InsertUser	1.39	2.60	1.33	0.45	17.97	4.03	0.41
PatchUser	1.43	0.67	10.56	0.44	19.55	4.49	0.41
GetUser	0.28	0.28	2.21	0.30	0.30	0.30	0.20
InsertUsers/10	1.48	2.21	1.47	1.00	21.59	4.51	0.92
InsertUsers/100	2.26	3.85	2.48	5.01	17.97	7.52	4.95
InsertUsers/1000	17.55	18.97	9.41	51.13	52.84	40.60	46.63
GetUsers/10	0.42	0.34	2.72	0.37	0.44	0.35	0.27
GetUsers/100	0.81	0.61	3.02	0.83	0.90	0.69	0.63
GetUsers/1000	3.57	2.89	6.28	5.20	4.81	10.23	3.85
DeleteUser	11.76	2.26	31.15	0.45	12.27	2.40	0.37
InsertItem	13.30	2.64	4.61	0.89	38.12	8.70	1.22
PatchItem	12.47	3.18	24.69	1.30	41.50	9.47	1.80
GetItem	0.29	0.30	3.15	0.21	0.37	0.37	0.26
InsertItems/10	10.72	2.98	5.02	1.09	18.32	7.57	2.70
InsertItems/100	11.42	5.19	6.83	4.50	46.67	15.04	13.01
InsertItems/1000	38.76	29.40	20.56	39.57	454.27	91.44	101.66
GetItems/10	0.39	0.40	3.64	0.32	0.49	0.48	0.31
GetItems/100	0.92	0.95	4.10	0.85	1.22	1.15	0.87
GetItems/1000	5.84	13.93	10.57	6.65	7.71	5.85	6.42
DeleteItem	10.42	3.39	27.15	1.06	25.00	6.15	1.73
InsertCategory	11.23	2.34	4.81	0.57	21.02	4.92	0.71
DeleteCategory	0.65	2.47	4.78	0.60	1.03	2.87	0.64
PutFeedback/10	10.09	3.08	5.67	2.28	31.26	8.33	2.65
PutFeedback/100	16.62	8.59	8.19	14.43	53.50	20.48	17.79
PutFeedback/1000	56.39	57.64	21.36	105.29	185.60	102.84	148.21
InsertFeedback/10	11.25	3.62	4.68	1.93	32.90	7.96	2.63
InsertFeedback/100	15.03	8.54	7.50	11.43	56.50	16.54	17.32
InsertFeedback/1000	62.52	58.02	23.98	103.05	152.69	121.93	153.70
GetFeedback/10	0.35	0.49	3.19	0.29	0.43	0.45	0.28
GetFeedback/100	0.63	0.88	3.56	0.70	0.85	0.82	0.70
GetFeedback/1000	3.11	2.98	9.79	4.56	4.26	2.97	4.78
GetUserItemFeedback	0.31	0.33	2.63	0.23	0.39	0.37	0.23
DeleteUserItemFeedback	10.10	2.00	12.80	0.24	10.93	2.25	0.28
GetUserFeedback	0.32	0.31	2.68	0.26	0.38	0.34	0.23
GetItemFeedback	0.31	0.31	2.50	0.22	0.38	0.35	0.22
GetRecommendCache/10	0.26	0.29	0.31	0.27	0.68	0.63	0.44
GetRecommendCache/100	0.50	0.48	0.60	0.49	3.05	1.30	1.25
GetRecommendCache/1000	2.48	2.48	3.20	2.67	119.33	13.80	8.25
RecommendFromOfflineCache/10	0.35	0.38	0.39	0.34	0.93	0.76	0.58
RecommendFromOfflineCache/100	0.59	0.60	0.73	0.58	1.98	1.49	1.43
RecommendFromOfflineCache/1000	2.83	2.84	3.73	2.87	14.69	5.11	7.89
RecommendFromLatest/10	0.69	0.71	3.20	0.61	1.47	1.23	0.91
RecommendFromLatest/100	1.04	0.93	3.89	1.13	2.82	1.69	2.21
RecommendFromLatest/1000	4.61	3.53	10.45	5.14	20.44	5.45	9.37
RecommendFromItemBased/10	2.21	2.21	5.86	1.94	6.75	5.10	3.79
RecommendFromItemBased/100	4.35	3.86	9.75	3.87	21.53	10.08	10.81
RecommendFromItemBased/1000	23.65	21.59	36.04	22.25	183.78	57.02	59.73

---

1. The first logo in the cell represents the database for data storage, and the second logo represents the database for cache storage. ↩︎