From c988b197dce69f175427048a88d55a19de8ebeb6 Mon Sep 17 00:00:00 2001
From: gifrerenom <lluis.gifre@cttc.es>
Date: Wed, 8 Nov 2023 18:37:37 +0000
Subject: [PATCH] Forecaster component:
- Updated TODO.txt
- added parse_csv.py script
---
parse_csv.py | 49 +++++++++++++++++++++++++++++++++
src/forecaster/service/TODO.txt | 31 ++++-----------------
2 files changed, 55 insertions(+), 25 deletions(-)
create mode 100644 parse_csv.py
diff --git a/parse_csv.py b/parse_csv.py
new file mode 100644
index 000000000..fc21c6a12
--- /dev/null
+++ b/parse_csv.py
@@ -0,0 +1,49 @@
+import calendar, pandas
+from datetime import datetime, timezone
+
+def time_datetime_to_int(dt_time : datetime) -> int:
+ return int(calendar.timegm(dt_time.timetuple()))
+
+def time_datetime_to_float(dt_time : datetime) -> float:
+ return time_datetime_to_int(dt_time) + (dt_time.microsecond / 1.e6)
+
+def time_utc_now_to_datetime() -> datetime:
+ return datetime.now(tz=timezone.utc)
+
+def time_utc_now_to_float() -> float:
+ return time_datetime_to_float(time_utc_now_to_datetime())
+
+def read_csv(csv_file : str) -> pandas.DataFrame:
+ df = pandas.read_csv(csv_file)
+
+ if 'dataset.csv' in csv_file:
+ df.rename(columns={'linkid': 'link_id', 'ds': 'timestamp', 'y': 'used_capacity_gbps'}, inplace=True)
+ df[['source', 'destination']] = df['link_id'].str.split('_', expand=True)
+ elif 'dataset2.csv' in csv_file:
+ df.drop(columns=['Unnamed: 0'], inplace=True)
+ df.rename(columns={
+ 'target': 'destination', 'id': 'link_id', 'ds': 'timestamp', 'demandValue': 'used_capacity_gbps'
+ }, inplace=True)
+
+ df['timestamp'] = pandas.to_datetime(df['timestamp'])
+ max_timestamp = time_datetime_to_int(df['timestamp'].max())
+ now_timestamp = time_datetime_to_int(datetime.now(tz=timezone.utc))
+ df['timestamp'] = df['timestamp'] + pandas.offsets.Second(now_timestamp - max_timestamp)
+ df.sort_values('timestamp', ascending=True, inplace=True)
+ return df
+
+CSV_FILE = 'data/forecaster_data/dataset.csv'
+#CSV_FILE = 'data/forecaster_data/dataset2.csv'
+
+def main():
+ df = read_csv(CSV_FILE)
+ #print(df)
+
+ #print(df.groupby(['source', 'destination']).count())
+
+ links = df.groupby(['source', 'destination']).all()
+ for link in links:
+ print(link)
+
+if __name__ == '__main__':
+ main()
diff --git a/src/forecaster/service/TODO.txt b/src/forecaster/service/TODO.txt
index eb3cf3652..95d3fba03 100644
--- a/src/forecaster/service/TODO.txt
+++ b/src/forecaster/service/TODO.txt
@@ -1,29 +1,10 @@
-Pseudocode for RPC method `ComputeTopologyForecast`:
-```python
- # Setting to configure the ratio between requested forecast and amount of historical data to be used for the forecast.
- # E.g., if forecast window is 1 week, compute forecast based on 10 weeks of historical data.
- FORECAST_TO_HISTORY_RATIO = 10
+test case:
+- instantiate Context and Monitoring
+- populate topology
+- populate monitoring timeseries with link utilizations
+- run test for link
+- run test for topology
- history_window_seconds = FORECAST_TO_HISTORY_RATIO * request.forecast_window_seconds
-
- forecast_reply = ForecastTopologyCapacityReply()
-
- topology = context_client.GetTopology(topology_id)
- for link_id in topology.link_ids:
- link = context_client.GetLink(link_id)
-
- used_capacity_history_gbps = monitoring_client.GetKPIValue(link_id, KPI.LinkUsedCapacity, window=history_window_seconds)
- forecast_used_capacity_gbps = compute_forecast(used_capacity_history_gbps, forecast_window_seconds)
-
- forecast_reply.link_capacities.append(ForecastLinkCapacityReply(
- link_id=link_id,
- total_capacity_gbps=link.total_capacity_gbps,
- current_used_capacity_gbps=link.used_capacity_gbps,
- forecast_used_capacity_gbps=forecast_used_capacity_gbps
- ))
-
- return forecast_reply
-```
## PathComp Impact
After retrieving the topology, if the service has a duration constraint configured, the PathComp component should interrogate the Forecaster and request a topology forecast according to the requested duration of the service. The computed link capacity forecast should be used as link capacity in path computations.
--
GitLab