You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: paper_reproduced/scripts/README.md
+27-27Lines changed: 27 additions & 27 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -3,26 +3,26 @@
3
3
4
4
All runs were executed 10 times to calculate the average results for each method.
5
5
6
-
## 📊 Fig 9 and Fig 10 (cuJSON vs CPU Methods)
6
+
## 📊 Fig 9 and Fig 10 (cujson vs CPU Methods)
7
7
8
8
To run each method separately and get the results for each dataset (executed 10 times), use the following commands:
9
9
10
-
-**cuJSON**
10
+
-**cujson**
11
11
```bash
12
12
./run_cujson_fig9.sh
13
13
```
14
14
15
-
-**simdJSON**
15
+
-**simdjson**
16
16
```bash
17
17
./run_simdjson_fig9.sh
18
18
```
19
19
20
-
-**RapidJSON**
20
+
-**rapidjson**
21
21
```bash
22
22
./run_rapidjson_fig9.sh
23
23
```
24
24
25
-
-**Pison**
25
+
-**pison**
26
26
```bash
27
27
./run_pison_fig9.sh
28
28
```
@@ -43,18 +43,18 @@ run_all_fig9.sh
43
43
44
44
---
45
45
46
-
## 📈 Fig 11 (cuJSON vs GPU Methods)
46
+
## 📈 Fig 11 (cujson vs GPU Methods)
47
47
48
48
To run each method separately for Fig 11:
49
49
50
-
-**GPJSON**
50
+
-**gpjson**
51
51
```bash
52
52
./run_gpjson_fig11.sh
53
53
```
54
54
55
-
-**cuDF**
55
+
-**cudf**
56
56
```bash
57
-
./run_cuDF_fig11.sh
57
+
./run_cudf_fig11.sh
58
58
```
59
59
60
60
If you'd like to run both methods together, use the following:
@@ -76,17 +76,17 @@ run_all_fig11.sh
76
76
77
77
To report peak GPU memory usage for each method:
78
78
79
-
-**cuJSON**
79
+
-**cujson**
80
80
```bash
81
81
./run_cujson_fig12.sh
82
82
```
83
83
84
-
-**GPJSON**
84
+
-**gpjson**
85
85
```bash
86
86
./run_gpjson_fig12.sh
87
87
```
88
88
89
-
-**cuDF**
89
+
-**cudf**
90
90
```bash
91
91
./run_cudf_fig12.sh
92
92
```
@@ -108,9 +108,9 @@ To run all methods together:
108
108
109
109
---
110
110
111
-
## ⏱️ Fig 13 and Table 8 (cuJSON only)
111
+
## ⏱️ Fig 13 and Table 8 (cujson only)
112
112
113
-
For time-breakdown reporting with **cuJSON**, execute the following:
113
+
For time-breakdown reporting with **cujson**, execute the following:
114
114
115
115
```bash
116
116
./run_cujson_fig13.sh
@@ -129,32 +129,32 @@ For time-breakdown reporting with **cuJSON**, execute the following:
129
129
130
130
To run each method separately and report memory usage:
131
131
132
-
-**cuJSON**
132
+
-**cujson**
133
133
```bash
134
134
./run_cujson_fig14.sh
135
135
```
136
136
137
-
-**simdJSON**
137
+
-**simdjson**
138
138
```bash
139
139
./run_simdjson_fig14.sh
140
140
```
141
141
142
-
-**RapidJSON**
142
+
-**rapidjson**
143
143
```bash
144
144
./run_rapidjson_fig14.sh
145
145
```
146
146
147
-
-**Pison**
147
+
-**pison**
148
148
```bash
149
149
./run_pison_fig14.sh
150
150
```
151
151
152
-
-**cuDF/MetaJSON**
152
+
-**cudf/MetaJSON**
153
153
```bash
154
154
./run_cudf_fig14.sh
155
155
```
156
156
157
-
> **Note**: To compute results for **GPJSON**, the library requires modifications. This library lead to no output and it required to modify the source code after installation. However, Theoritcally, the best possible results of `GPJSON` will be as same as `pison`, but by modification we realize it is even worst than `pison`.
157
+
> **Note**: To compute results for **gpjson**, the library requires modifications. This library lead to no output and it required to modify the source code after installation. However, Theoritcally, the best possible results of `gpjson` will be as same as `pison`, but by modification we realize it is even worst than `pison`.
158
158
159
159
To run all methods together:
160
160
@@ -179,7 +179,7 @@ To run all methods together:
179
179
180
180
To report the average time for running the queries, execute the scripts for each method:
181
181
182
-
-**cuJSON**
182
+
-**cujson**
183
183
```bash
184
184
./run_cujson_fig15.sh
185
185
```
@@ -189,12 +189,12 @@ To report the average time for running the queries, execute the scripts for each
189
189
./run_simdjson_fig15.sh
190
190
```
191
191
192
-
-**Pison**
192
+
-**pison**
193
193
```bash
194
194
./run_pison_fig15.sh
195
195
```
196
196
197
-
-**RapidJSON**
197
+
-**rapidjson**
198
198
```bash
199
199
./run_rapidjson_fig15.sh
200
200
```
@@ -219,14 +219,14 @@ To run all methods together:
219
219
220
220
### Middle and Right: Modifying JSON Files
221
221
222
-
For these sections, you will need to modify the JSON file to contain only one record to compute the query time. Library modifications are required for **GPJSON** and **cuDF**, so ensure you install the libraries and use them accordingly.
222
+
For these sections, you will need to modify the JSON file to contain only one record to compute the query time. Library modifications are required for **gpjson** and **cudf**, so ensure you install the libraries and use them accordingly.
223
223
224
224
225
225
---
226
226
227
-
## 🧑💻 Fig 16: cuJSON Scalability
227
+
## 🧑💻 Fig 16: cujson Scalability
228
228
229
-
To run **cuJSON** scalability tests:
229
+
To run **cujson** scalability tests:
230
230
231
231
1. First, download the scalability data.
232
232
2. Then, execute the following:
@@ -244,7 +244,7 @@ To run **cuJSON** scalability tests:
244
244
## General Notes:
245
245
246
246
- All scripts assume you have the necessary dependencies installed.
247
-
- For `GPJSON` after you install the required library files, you have to add your keys in the scripts: `scripts/run_gpjson_fig11.sh` and `scripts/run_gpjson_fig12.sh`.
247
+
- For `gpjson` after you install the required library files, you have to add your keys in the scripts: `scripts/run_gpjson_fig11.sh` and `scripts/run_gpjson_fig12.sh`.
248
248
- The `scripts/results` folder will contain all output files, categorized by method and csvs.
249
249
- For specific modifications or troubleshooting, refer to the individual script files, or readme of each relarted works for more details.
250
250
- Make sure to download all the datasets can be downloaded from https://drive.google.com/drive/folders/1PkDEy0zWOkVREfL7VuINI-m9wJe45P2Q?usp=sharing and placed into the `dataset` folder. `scabality` folder must place exactly like what it is in the `dataset` folder for proper experiment.
0 commit comments