In [ ]:
# Catch-up cell. Run if you did not do previous notebooks of this sequence
!wget http://kdd.ics.uci.edu/databases/kddcup99/kddcup.data_10_percent.gz
BUCKET='cloud-training-demos-ml' # CHANGE
!gsutil cp kdd* gs://$BUCKET/
In [ ]:
!bq mk sparktobq
In [ ]:
BUCKET='cloud-training-demos-ml' # CHANGE
!bq --location=US load --autodetect --source_format=CSV sparktobq.kdd_cup_raw gs://$BUCKET/kddcup.data_10_percent.gz
In [ ]:
%%bigquery
SELECT * FROM sparktobq.kdd_cup_raw LIMIT 5
Ooops. There are no column headers. Let's fix this.
In [ ]:
%%bigquery
CREATE OR REPLACE TABLE sparktobq.kdd_cup AS
SELECT
int64_field_0 AS duration,
string_field_1 AS protocol_type,
string_field_2 AS service,
string_field_3 AS flag,
int64_field_4 AS src_bytes,
int64_field_5 AS dst_bytes,
int64_field_6 AS wrong_fragment,
int64_field_7 AS urgent,
int64_field_8 AS hot,
int64_field_9 AS num_failed_logins,
int64_field_11 AS num_compromised,
int64_field_13 AS su_attempted,
int64_field_14 AS num_root,
int64_field_15 AS num_file_creations,
string_field_41 AS label
FROM
sparktobq.kdd_cup_raw
In [ ]:
%%bigquery
SELECT * FROM sparktobq.kdd_cup LIMIT 5
In [ ]:
%%bigquery connections_by_protocol
SELECT COUNT(*) AS count
FROM sparktobq.kdd_cup
GROUP BY protocol_type
ORDER by count ASC
In [ ]:
connections_by_protocol
In [ ]:
%%bigquery attack_stats
SELECT
protocol_type,
CASE label
WHEN 'normal.' THEN 'no attack'
ELSE 'attack'
END AS state,
COUNT(*) as total_freq,
ROUND(AVG(src_bytes), 2) as mean_src_bytes,
ROUND(AVG(dst_bytes), 2) as mean_dst_bytes,
ROUND(AVG(duration), 2) as mean_duration,
SUM(num_failed_logins) as total_failed_logins,
SUM(num_compromised) as total_compromised,
SUM(num_file_creations) as total_file_creations,
SUM(su_attempted) as total_root_attempts,
SUM(num_root) as total_root_acceses
FROM sparktobq.kdd_cup
GROUP BY protocol_type, state
ORDER BY 3 DESC
In [ ]:
%matplotlib inline
ax = attack_stats.plot.bar(x='protocol_type', subplots=True, figsize=(10,25))
In [ ]:
import google.cloud.storage as gcs
# save locally
ax[0].get_figure().savefig('report.png');
connections_by_protocol.to_csv("connections_by_protocol.csv")
# upload to GCS
bucket = gcs.Client().get_bucket(BUCKET)
for blob in bucket.list_blobs(prefix='sparktobq/'):
blob.delete()
for fname in ['report.png', 'connections_by_protocol.csv']:
bucket.blob('sparktobq/{}'.format(fname)).upload_from_filename(fname)
Copyright 2019 Google Inc. Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.