Skip to content

Commit 3347fe1

Browse files
committed
Update artifact and add configs and plots
1 parent 4e4ad8c commit 3347fe1

39 files changed

+625
-1
lines changed

.gitignore

Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -15,3 +15,7 @@ Cargo.lock
1515

1616
# Ignore config files
1717
**/*.json
18+
19+
# Figs
20+
*.pdf
21+
*.png

README.md

Lines changed: 18 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -146,7 +146,7 @@ mode = "plain_metrics"
146146
# ...
147147
```
148148

149-
#### Plain Metrics with Prios: Aggregators
149+
#### Plain Metrics with Prio: Aggregators
150150
Run the aggregators in two separate shells. They will wait and be ready to
151151
process client requests.
152152
```bash
@@ -172,6 +172,23 @@ This branch can do Plain Heavy Hitters by setting the histogram size to 1, but a
172172
more efficient implementation uses the `Count` circuit and is in the [`Count`
173173
branch](https://github.com/TrustworthyComputing/mastic/tree/Count).
174174

175+
## Troubleshooting
176+
Mastic relies on the [tarpc](https://github.com/google/tarpc) library which has
177+
a limit on the size of the RPC messages. As such, you might see an error similar
178+
to the following:
179+
```shell
180+
thread 'main' panicked at src/bin/driver.rs:335:
181+
called `Result::unwrap()` on an `Err` value: Disconnected
182+
```
183+
which is caused by the RPC batch sizes.
184+
185+
To fix this reduce the batch sizes of either the reports or the FLPs (or both).
186+
Note that this does not affect the online running time, but it affects the
187+
upload time from the `driver` to the Mastic servers.
188+
```toml
189+
add_report_share_batch_size = 1000
190+
query_flp_batch_size = 100000
191+
```
175192

176193
## Disclaimer
177194

Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -301,3 +301,6 @@ zipf_exponent = 1.03
301301
```
302302
etc. These parameters are sufficient to reproduce all our results -- all our
303303
experiments in the paper specify the parameters used.
304+
305+
To reproduce our experiments, use the configs from the [configs](./configs/)
306+
directory and the scripts from the [plots](../plots/) directory.
Lines changed: 14 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,14 @@
1+
# For more efficiency, use the "count" branch in case m = 1.
2+
3+
mode.weighted_heavy_hitters.threshold = 0.01
4+
5+
data_bits = 128
6+
hist_buckets = 1
7+
8+
server_0 = "0.0.0.0:8000"
9+
server_1 = "0.0.0.0:8001"
10+
11+
add_report_share_batch_size = 100
12+
query_flp_batch_size = 100000
13+
zipf_unique_buckets = 1000
14+
zipf_exponent = 1.03
Lines changed: 14 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,14 @@
1+
# For more efficiency, use the "count" branch in case m = 1.
2+
3+
mode.weighted_heavy_hitters.threshold = 0.01
4+
5+
data_bits = 256
6+
hist_buckets = 1
7+
8+
server_0 = "0.0.0.0:8000"
9+
server_1 = "0.0.0.0:8001"
10+
11+
add_report_share_batch_size = 100
12+
query_flp_batch_size = 100000
13+
zipf_unique_buckets = 1000
14+
zipf_exponent = 1.03
Lines changed: 14 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,14 @@
1+
# For more efficiency, use the "count" branch in case m = 1.
2+
3+
mode.weighted_heavy_hitters.threshold = 0.01
4+
5+
data_bits = 64
6+
hist_buckets = 1
7+
8+
server_0 = "0.0.0.0:8000"
9+
server_1 = "0.0.0.0:8001"
10+
11+
add_report_share_batch_size = 100
12+
query_flp_batch_size = 100000
13+
zipf_unique_buckets = 1000
14+
zipf_exponent = 1.03
Lines changed: 12 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,12 @@
1+
mode.weighted_heavy_hitters.threshold = 0.01
2+
3+
data_bits = 128
4+
hist_buckets = 9
5+
6+
server_0 = "0.0.0.0:8000"
7+
server_1 = "0.0.0.0:8001"
8+
9+
add_report_share_batch_size = 100
10+
query_flp_batch_size = 100000
11+
zipf_unique_buckets = 1000
12+
zipf_exponent = 1.03
Lines changed: 12 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,12 @@
1+
mode.weighted_heavy_hitters.threshold = 0.01
2+
3+
data_bits = 256
4+
hist_buckets = 9
5+
6+
server_0 = "0.0.0.0:8000"
7+
server_1 = "0.0.0.0:8001"
8+
9+
add_report_share_batch_size = 100
10+
query_flp_batch_size = 100000
11+
zipf_unique_buckets = 1000
12+
zipf_exponent = 1.03
Lines changed: 12 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,12 @@
1+
mode.weighted_heavy_hitters.threshold = 0.01
2+
3+
data_bits = 64
4+
hist_buckets = 9
5+
6+
server_0 = "0.0.0.0:8000"
7+
server_1 = "0.0.0.0:8001"
8+
9+
add_report_share_batch_size = 100
10+
query_flp_batch_size = 100000
11+
zipf_unique_buckets = 1000
12+
zipf_exponent = 1.03
Lines changed: 12 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,12 @@
1+
mode.weighted_heavy_hitters.threshold = 0.01
2+
3+
data_bits = 128
4+
hist_buckets = 29
5+
6+
server_0 = "0.0.0.0:8000"
7+
server_1 = "0.0.0.0:8001"
8+
9+
add_report_share_batch_size = 100
10+
query_flp_batch_size = 100000
11+
zipf_unique_buckets = 1000
12+
zipf_exponent = 1.03

0 commit comments

Comments
 (0)