Community Articles
Find and share helpful community-sourced technical articles
Super Guru

Smart Stocks with FLaNK (NiFi, Kafka, Flink SQL)

TimothySpann_0-1608657207264.png
I would like to track stocks from IBM and Cloudera frequently during the day using Apache NiFi to read the REST API. After that I have some Streaming Analytics to perform with Apache Flink SQL and I also want permanent fast storage in Apache Kudu queried with Apache Impala.
 
Let's build that application cloud native in seconds in AWS or Azure.
To script Loading Schemas, Tables, Alerts, see scripts/setup.sh:
  • Kafka Topic
  • Kafka Schema
  • Kudu Table
  • Flink Prep
  • Flink SQL Client Run
  • Flink SQL Client Configuration
Once our automated admin has built our Cloud environment and populated it with the goodness of our app, we can bring out continuous SQL.
 
TimothySpann_1-1608657207251.jpeg

If you know your data, build a schema, share to the registry

TimothySpann_2-1608657207091.png

 

TimothySpann_3-1608657207179.png
One unique thing we added was a default value in our Avro schema and making it a logicalType for timestamp-millis. This is helpful for Flink SQL timestamp related queries.
 
{ "name" : "dt", "type" : ["long"], "default": 1, "logicalType": "timestamp-millis"}

You can see the entire schema here:

https://raw.githubusercontent.com/tspannhw/SmartStocks/main/stocks.avsc

We will also want a topic for Stock Alerts that we will create later with Flink SQL., so let's define a schema for that as well.

https://raw.githubusercontent.com/tspannhw/SmartStocks/main/stockalerts.avsc

For our data today, we will use AVRO data with AVRO schemas for use inside Kafka topics and whoever will consume it.

How to Build a Smart Stock DataFlow in X Easy Steps

TimothySpann_4-1608657207262.png

TimothySpann_5-1608657207259.png
  1. Retrieve data from source (example: InvokeHTTP against SSL REST Feed - say TwelveData) with a schedule
  2. Set a Schema Name (UpdateAttribute)
  3. ForkRecord: We use this to split out records from the header (/values) using RecordPath syntax.
  4. QueryRecord: Convert type and manipulate data with SQL. We aren't doing anything in this one, but this is an option to change fields, add fields, etc.
  5. UpdateRecord: This first one I am setting some fields in the record from attributes and adding a current timestamp. I also reformat by timestamp for conversion.
  6. UpdateRecord: I am making dt make numeric UNIX timestamp.
  7. UpdateRecord: I am making datetime my formatted String date time.
  8. (LookupRecord): I don't have this step yet as I don't have an internal record for this company in my Real-Time Data Mart. Apache Kudu architecture in a CDP public cloud deployment. I will probably add this step to augment or check my data.
  9. (ValidateRecord): For less reliable data sources, I may want to validate my data against our schema, otherwise we will get warnings or errors.
  10. PublishKafkaRecord_2_0: Convert from JSON to AVRO, send to our Kafka topic with headers, including reference to the correct schema stocks and it's version 1.0.
TimothySpann_6-1608657207077.png

 

TimothySpann_7-1608657207149.png
TimothySpann_8-1608657207126.png
TimothySpann_9-1608657207093.png
TimothySpann_10-1608657207096.png
TimothySpann_11-1608657207147.png
Now that we are streaming our data to Kafka topics, we can utilize it in Flink SQL Continuous SQL applications, NiFi applications, Spark 3 applications and more. So in this case, CFM NiFi is our Producer and we will have CFM NiFi and CSA Flink SQL as Kafka Consumers.
We can see what our data looks like in the new cleaned up format with all the fields we need.
 
TimothySpann_12-1608657207248.png

Viewing, Monitoring, Checking and Alerting On Our Streaming Data in Kafka

Cloudera Streams Messaging Manager solves all of these difficult problems from one easy to use pre-integrated UI. It is pre-wired into my Kafka Datahubs and secured with SDX.
 
TimothySpann_13-1608657207148.png

 

TimothySpann_14-1608657207050.png

 

TimothySpann_15-1608657207180.png
I can see my AVRO data with associated stocks schema in the topic, ready to be consumed. I can then monitor who is consuming how much, and if there is a lag or latency.

How to Store Our Streaming Data to Our Real-Time DataMart in the Cloud

TimothySpann_16-1608657207224.png
Consume stocks AVRO data with stocks schema then write to our Real-Time Data Mart in Cloudera Data Platform powered by Apache Impala and Apache Kudu. If something failed or could not connect, let's retry 3 times.
 
We use a parameter for our 3+ Kafka brokers with port. We could also have parameters for topic names and consumer name. We read from stocks table, which uses stocks schema that is referenced in Kafka header automatically ready by NiFi. When we sent message to Kafka, Nifi passed on our schema name through schema.name attribute in NiFi. As we can see, it was schema attached Avro, so we use that Reader and convert to simple JSON with that schema.
TimothySpann_17-1608657207173.png
TimothySpann_18-1608657207141.png
Writing to our Cloud Native Real-Time Data Mart could not be simpler, we reference the table stocks we have created and have permissions to and use the JSON reader. I like UPSERT since it handles INSERT AND UPDATE.
 
First we need to create our Kudu table in either Apache Hue from CDP or from the command line scripted. 
Example:
impala-shell -i edge2ai-1.dim.local -d default -f /opt/demo/sql/kudu.sql

 

 

 

CREATE TABLE stocks
(
  uuid STRING,
  `datetime` STRING,
  `symbol` STRING, 
  `open` STRING, 
  `close` STRING,
  `high` STRING,
  `volume` STRING,
  `ts` TIMESTAMP,
  `dt`  TIMESTAMP,
  `low` STRING,
PRIMARY KEY (uuid,`datetime`) ) 
PARTITION BY HASH PARTITIONS 4 
STORED AS KUDU TBLPROPERTIES ('kudu.num_tablet_replicas' = '1');
TimothySpann_19-1608657207220.png

Using Apache Hue integrated in CDP, I can examine my Real-Time Data Mart table and then query my table.

TimothySpann_20-1608657207112.png
My data is now ready for reports, dashboards, applications, notebooks, web applications, mobile apps and machine learning.
TimothySpann_21-1608657207235.png
I can now spin up a Cloudera Visual Application on this table in a few seconds.
 
TimothySpann_22-1608657207239.png
Now we can build our streaming analytics application in Flink.

How to Build a Smart Stock Streaming Analytics in X Easy Steps

I can connect to Flink SQL from the command line Flink SQL Client to start exploring my Kafka and Kudu data, create temporary tables and launch some applications (Insert statements). The environment lets me see all the different catalogs available, including registry (Cloudera Cloud Schema Registry), Hive (Cloud Native Database table) and Kudu (Cloudera Real-Time Cloud Data Mart) tables.

TimothySpann_23-1608657207255.png
TimothySpann_24-1608657207200.png
 
TimothySpann_25-1608657207242.png
TimothySpann_26-1608657207252.png
TimothySpann_27-1608657207214.png
TimothySpann_28-1608657207215.png
TimothySpann_29-1608657207266.png
Run Flink SQL Client
It's a two step process; first setup a YARN session. You may need to add your Kerberos credentials.
 
flink-yarn-session -tm 2048 -s 2 -d
Then launch the command line SQL Client.
 
flink-sql-client embedded -e sql-env.yaml
Refer to:

Run Flink SQL

Cross Catalog Query to Stocks Kafka Topic

select * from registry.default_database.stocks;

Cross Catalog Query to Stocks Kudu/Impala Table

select * from kudu.default_database.impala::default.stocks;

Default Catalog

use catalog default_catalog;

 

CREATE TABLE stockEvents ( symbol STRING, uuid STRING, ts BIGINT, dt BIGINT, datetime STRING, open STRING, close STRING, high STRING, volume STRING, low STRING,
event_time AS CAST(from_unixtime(floor(ts/1000)) AS TIMESTAMP(3)),
WATERMARK FOR event_time AS event_time - INTERVAL '5' SECOND )
WITH ( 'connector.type' = 'kafka', 'connector.version' = 'universal',
'connector.topic' = 'stocks',
'connector.startup-mode' = 'earliest-offset',
'connector.properties.bootstrap.servers' = 'edge2ai-1.dim.local:9092',
'format.type' = 'registry',
'format.registry.properties.schema.registry.url' = 'http://edge2ai-1.dim.local:7788/api/v1' );

 

 

show tables;
 
Flink SQL> describe stockEvents;
 

root |-- symbol: STRING |-- uuid: STRING |-- ts: BIGINT |-- dt: BIGINT |-- datetime: STRING |-- open: STRING |-- close: STRING |-- high: STRING |-- volume: STRING |-- low: STRING |-- event_time: TIMESTAMP(3) AS CAST(FROM_UNIXTIME(FLOOR(ts / 1000)) AS TIMESTAMP(3)) |-- WATERMARK FOR event_time AS event_time - INTERVAL '5' SECOND

 

We added a watermark and event time pulled from our timestamp.

Simple Select All Query

select * from default_catalog.default_database.stockEvents;
 
We can do some interesting queries against this table we created.

Tumbling Window

SELECT symbol,
TUMBLE_START(event_time, INTERVAL '1' MINUTE) as tumbleStart,
TUMBLE_END(event_time, INTERVAL '1' MINUTE) as tumbleEnd,
AVG(CAST(high as DOUBLE)) as avgHigh
FROM stockEvents
WHERE symbol is not null
GROUP BY TUMBLE(event_time, INTERVAL '1' MINUTE), symbol;

 

Top 3

SELECT *
FROM
( SELECT * , ROW_NUMBER() OVER
( PARTITION BY window_start ORDER BY num_stocks desc ) AS rownum
FROM (
SELECT TUMBLE_START(event_time, INTERVAL '10' MINUTE) AS window_start,
symbol,
COUNT(*) AS num_stocks
FROM stockEvents
GROUP BY symbol,
TUMBLE(event_time, INTERVAL '10' MINUTE) ) )
WHERE rownum <=3;

 

Stock Alerts

INSERT INTO stockalerts
/*+ OPTIONS('sink.partitioner'='round-robin') */
SELECT CAST(symbol as STRING) symbol,
CAST(uuid as STRING) uuid, ts, dt, open, close, high, volume, low,
datetime, 'new-high' message, 'nh' alertcode, CAST(CURRENT_TIMESTAMP AS BIGINT) alerttime FROM stocks st
WHERE symbol is not null
AND symbol <> 'null'
AND trim(symbol) <> ''
AND CAST(close as DOUBLE) > 11;

Monitoring Flink Jobs

Using the CSA Flink Global Dashboard, I can see all my Flink jobs running, including SQL Client jobs, disconnected Flink SQL inserts and deployed Flink applications.
 
TimothySpann_30-1608657207194.png

 

TimothySpann_31-1608657207196.png

 

TimothySpann_32-1608657207195.png

 

TimothySpann_33-1608657207222.png

 

TimothySpann_34-1608657207163.png

 

TimothySpann_35-1608657207177.png

 

TimothySpann_36-1608657207210.png
We can also see the data populated in the stockalerts topic. We can run a Flink SQL, Spark 3, NiFi or other applications against this data to handle alerts. That may be the next application - I may send those alerts to Iphone messages, Slack messages, a database table and a websockets app.
TimothySpann_37-1608657207174.png

Data Lineage and Governance

We all know that NiFi has deep data lineage that can be pushed or pulled using REST, Reporting Tasks or CLI to use in audits, metrics and tracking. If I want all the governance data for my entire streaming pipeline, I will use Apache Atlas that is prewired as part of SDX in my Cloud Data Platform.
 
TimothySpann_38-1608657207201.png

References

TimothySpann_39-1608657207267.jpeg
448 Views
Tags (3)
Comments

Amazing work here sir!  

Don't have an account?
Version history
Last update:
‎12-22-2020 08:36 PM
Updated by:
Top Kudoed Authors