Watermarking by key in Spark structured streaming - apache-spark

I have data coming in on Kafka from IoT devices. The timestamps of the sensor data of these devices are often not in sync due to network congestion, device being out of range, etc.
We have to write streaming jobs to aggregate sensor values over a window of time for each device independently. With the groupby with watermark operation, we lose the data of all devices that lag behind device with latest timestamp.
Is there any way that the watermarks could be applied separately to each device based on the latest timestamp for that device, and not the latest timestamp across all devices?
We cannot keep a large lag as the device could be out of range for days. We cannot run an individual query for each device as the number of devices is high.
Would it be achievable using flatMapGroupsWithState? Or is this something that cannot be achieved with Spark Structured Streaming at all?

I think instead of watermarking by event timestamp (which could be lagging behind as you said), you could apply a watermark over the processing timestamp (i.e. the time when you process the data in your Spark job). I faced a very similar problem in a recent project I was working on and that's how I solved it.
Example:
val dfWithWatermark = df
.withColumn("processingTimestamp", current_timestamp())
.withWatermark("processingTimestamp", "1 day")
// ... use dfWithWatermark to do aggregations etc
This will keep a state over 1 day of your IoT data, no matter what the timestamp of the data is that you're receiving.
There are some limitations to this of course, for example if there are devices that send data in intervals larger than your watermark. But to figure out a solution for this you'd have to be more specific with your requirements.
By using flatMapGroupsWithState you can be very specific with your state keeping, but I don't think it's really necessary in your case.
Edit: if you however decide to go with flatMapGroupsWithState, you can use different timeouts per device group by calling state.setTimeoutDuration() with different intervals, depending on the type of device you process. This way you can be very specific with the state keeping.

Related

Kappa architecture: when insert to batch/analytic serving layer happens

As you know, Kappa architecture is some kind of simplification of Lambda architecture. Kappa doesn't need batch layer, instead speed layer have to guarantee computation precision and enough throughput (more parallelism/resources) on historical data re-computation.
Still Kappa architecture requires two serving layers in case when you need to do analytic based on historical data. For example, data that have age < 2 weeks are stored at Redis (streaming serving layer), while all older data are stored somewhere at HBase (batch serving layer).
When (due to Kappa architecture) I have to insert data to batch serving layer?
If streaming layer inserts data immidiately to both batch & stream serving layers - than how about late data arrival? Or streaming layer should backup speed serving layer to batch serving layer on regular basis?
Example: let say source of data is Kafka, data are processed by Spark Structured Streaming or Flink, sinks are Redis and HBase. When write to Redis & HBase should happen?
If we perform stream processing, we want to make sure that output data is firstly made available as a data stream. In your example that means we write to Kafka as a primary sink.
Now you have two options:
have secondary jobs that reads from that Kafka topic and writes to Redis and HBase. That is the Kafka way, in that Kafka Streams does not support writing directly to any of these systems and you set up a Kafka connect job. These secondary jobs can then be tailored to the specific sinks, but they add additional operations overhead. (That's a bit of the backup option that you mentioned).
with Spark and Flink you also have the option to have secondary sinks directly in your job. You may add additional processing steps to transform the Kafka output into a more suitable form for the sink, but you are more limited when configuring the job. For example in Flink, you need to use the same checkpointing settings for the Kafka sink and the Redis/HBase sink. Nevertheless, if the settings work out, you just need to run one streaming job instead of 2 or 3.
Late events
Now the question is what to do with late data. The best solution is to let the framework handle that through watermarks. That is, data is only committed at all sinks, when the framework is sure that no late data arrives. If that doesn't work out because you really need to process late events even if they arrive much, much later and still want to have temporary results, you have to use update events.
Update events
(as requested by the OP, I will add more details to the update events)
In Kafka Streams, elements are emitted through a continuous refinement mechanism by default. That means, windowed aggregations emit results as soon as they have any valid data point and update that result while receiving new data. Thus, any late event is processed and yield an updated result. While this approach nicely lowers the burden to users, as they do not need to understand watermarks, it has some severe short-comings that led the Kafka Streams developers to add Suppression in 2.1 and onward.
The main issue is that it poses quite big challenges to downward users to process intermediate results as also explained in the article about Suppression. If it's not obvious if a result is temporary or "final" (in the sense that all expected events have been processed) then many applications are much harder to implement. In particular, windowing operations need to be replicated on consumer side to get the "final" value.
Another issue is that the data volume is blown up. If you'd have a strong aggregation factor, using watermark-based emission will reduce your data volume heavily after the first operation. However, continuous refinement will add a constant volume factor as each record triggers a new (intermediate) record for all intermediate steps.
Lastly, and particularly interesting for you is how to offload data to external systems if you have update events. Ideally, you would offload the data with some time lag continuously or periodically. That approach simulates the watermark-based emission again on consumer side.
Mixing the options
It's possible to use watermarks for the initial emission and then use update events for late events. The volume is then reduced for all "on-time" events. For example, Flink offers allowed lateness to make windows trigger again for late events.
This setup makes offloading data much easier as data only needs to be re-emitted to the external systems if a late event actually happened. The system should be tweaked that a late event is a rare case though.

Transparent Streaming & Batch processing

I'm still quite new to the world of stream and batch processing and trying to understnad concepts and speach. It is admitedly very possible that the answer to my question well known, easy to find or even answered a hundred times here at SO, but I was not able to find it.
The background:
I am working in a big scientific project (nuclear fusion research), and we are producing tons of measurement data during experiment runs. Those data are mostly streams of samples tagged with a nanosecond timestamp, where samples can be anything from a single by ADC value, via an array of such, via deeply structured data (with up to hundreds of entries from 1 bit booleans to 64bit double precision floats) to raw HD video frames or even string text messages. If I understand the common terminologies right, I would regard our data as "tabular data", for the most part.
We are working with mostly selfmade software solutions from data acquisition over simple online (streaming) analysis (like scaling, subsampling and such) to our own data sotrage, management and access facilities.
In view of the scale of the operation and the effort for maintaining all those implementations, we are investigating the possibilities to use standard frameworks and tools for more of our tasks.
My question:
In particular at this stage, we are facing the need for more and more sofisticated (automated and manual) data analytics on live/online/realtime data as well as "after the fact" offline/batch analytics of "historic" data. In this endavor, I am trying to understand if and how existing analytics frameworks like Spark, Flink, Storm etc. (possibly supported by message queues like Kafka, Pulsar,...) can support a scenario, where
data is flowing/streamed into the platform/framework, attached an identifier like a URL or an ID or such
the platform interacts with integrated or external storage to persist the streaming data (for years), associated with the identifier
analytics processes can now transparently query/analyse data addressed by an identifier and an arbitrary (open or closed) time window, and the framework suplies data batches/samples for the analysis either from backend storage or coming in live from data acquisition
Simply streaming the online data into storage and querying from there seems no option as we need both raw and analysed data for live monitoring and realtime feedback control of the experiment.
Also, letting the user query either a live input signal or a historic batch from storage differently would not be ideal, as our physicists mostly are no data scientists and we would like to keep such "technicalities" away from them and idealy the exact same algorithms should be used for analysing new real time data and old stored data from previous experiments.
Sitenotes:
we are talking about peek data loads in the range of 10th of gigabits per second coming in bursts of increasing length of seconds up to minutes - could this be handled by the candidates?
we are using timestamps in nanosecond resolution, even thinking about pico - this poses some limitations on the list of possible candidates if I unserstand correctly?
I would be very greatfull if anyone would be able to understand my question and to shed some light on the topic for me :-)
Many Thanks and kind regards,
Beppo
I don't think anyone can say "yes, framework X can definitely handle your workload", because it depends a lot on what you need out of your message processing, e.g. regarding messaging reliability, and how your data streams can be partitioned.
You may be interested in BenchmarkingDistributedStreamProcessingEngines. The paper is using versions of Storm/Flink/Spark that are a few years old (looks like they were released in 2016), but maybe the authors would be willing to let you use their benchmark to evaluate newer versions of the three frameworks?
A very common setup for streaming analytics is to go data source -> Kafka/Pulsar -> analytics framework -> long term data store. This decouples processing from data ingest, and lets you do stuff like reprocessing historical data as if it were new.
I think the first step for you should be to see if you can get the data volume you need through Kafka/Pulsar. Either generate a test set manually, or grab some data you think could be representative from your production environment, and see if you can put it through Kafka/Pulsar at the throughput/latency you need.
Remember to consider partitioning of your data. If some of your data streams could be processed independently (i.e. ordering doesn't matter), you should not be putting them in the same partitions. For example, there is probably no reason to mix sensor measurements and the video feed streams. If you can separate your data into independent streams, you are less likely to run into bottlenecks both in Kafka/Pulsar and the analytics framework. Separate data streams would also allow you to parallelize processing in the analytics framework much better, as you could run e.g. video feed and sensor processing on different machines.
Once you know whether you can get enough throughput through Kafka/Pulsar, you should write a small example for each of the 3 frameworks. To start, I would just receive and drop the data from Kafka/Pulsar, which should let you know early whether there's a bottleneck in the Kafka/Pulsar -> analytics path. After that, you can extend the example to do something interesting with the example data, e.g. do a bit of processing like what you might want to do in production.
You also need to consider which kinds of processing guarantees you need for your data streams. Generally you will pay a performance penalty for guaranteeing at-least-once or exactly-once processing. For some types of data (e.g. the video feed), it might be okay to occasionally lose messages. Once you decide on a needed guarantee, you can configure the analytics frameworks appropriately (e.g. disable acking in Storm), and try benchmarking on your test data.
Just to answer some of your questions more explicitly:
The live data analysis/monitoring use case sounds like it fits the Storm/Flink systems fairly well. Hooking it up to Kafka/Pulsar directly, and then doing whatever analytics you need sounds like it could work for you.
Reprocessing of historical data is going to depend on what kind of queries you need to do. If you simply need a time interval + id, you can likely do that with Kafka plus a filter or appropriate partitioning. Kafka lets you start processing at a specific timestamp, and if you data is partitioned by id or you filter it as the first step in your analytics, you could start at the provided timestamp and stop processing when you hit a message outside the time window. This only applies if the timestamp you're interested in is when the message was added to Kafka though. I also don't believe Kafka supports below-millisecond resolution on the timestamps it generates.
If you need to do more advanced queries (e.g. you need to look at timestamps generated by your sensors), you could look at using Cassandra or Elasticsearch or Solr as your permanent data store. You will also want to investigate how to get the data from those systems back into your analytics system. For example, I believe Spark ships with a connector for reading from Elasticsearch, while Elasticsearch provides a connector for Storm. You should check whether such a connector exists for your data store/analytics system combination, or be willing to write your own.
Edit: Elaborating to answer your comment.
I was not aware that Kafka or Pulsar supported timestamps specified by the user, but sure enough, they both do. I don't see that Pulsar supports sub-millisecond timestamps though?
The idea you describe can definitely be supported by Kafka.
What you need is the ability to start a Kafka/Pulsar client at a specific timestamp, and read forward. Pulsar doesn't seem to support this yet, but Kafka does.
You need to guarantee that when you write data into a partition, they arrive in order of timestamp. This means that you are not allowed to e.g. write first message 1 with timestamp 10, and then message 2 with timestamp 5.
If you can make sure you write messages in order to Kafka, the example you describe will work. Then you can say "Start at timestamp 'last night at midnight'", and Kafka will start there. As live data comes in, it will receive it and add it to the end of its log. When the consumer/analytics framework has read all the data from last midnight to current time, it will start waiting for new (live) data to arrive, and process it as it comes in. You can then write custom code in your analytics framework to make sure it stops processing when it reaches the first message with timestamp 'tomorrow night'.
With regard to support of sub-millisecond timestamps, I don't think Kafka or Pulsar will support it out of the box, but you can work around it reasonably easily. Just put the sub-millisecond timestamp in the message as a custom field. When you want to start at e.g. timestamp 9ms 10ns, you ask Kafka to start at 9ms, and use a filter in the analytics framework to drop all messages between 9ms and 9ms 10ns.
Allow me to add the following suggestions on how Apache Pulsar might help address some of your requirements. Food for thought as it were.
"data is flowing/streamed into the platform/framework, attached an identifier like a URL or an ID or such"
You might want to look at Pulsar Functions, which allows you to write simple functions (In Java or Python) that gets executed on each individual message that is published to a topic. They are ideal for this type of data augmentation use case.
the platform interacts with integrated or external storage to persist the streaming data (for years), associated with the identifier
Pulsar has recently added tiered-storage, that allows you to retain event streams in S3, Azure Blob Store, or Google Cloud storage. This would allow you to keep the data for years in a cheap and reliable data store
analytics processes can now transparently query/analyse data addressed by an identifier and an arbitrary (open or closed) time window, and the framework suplies data batches/samples for the analysis either from backend storage or coming in live from data acquisition
Apache Pulsar has also added integration with the Presto query engine, which would allow you to query the data over a given time period (including data from tiered-storage) and place it into a topic for processing.

How to apply rules that can be defined at runtime to streaming datasets?

Not sure the title is well suited to what I'm trying to achieve, so bear with me.
I'll start with defining my use case:
Many(say millions) IoT devices are sending data to my Spark stream. These devices are sending the current temperature level every 10 seconds.
The owner of all of these IoT devices has the ability to define a preset rules, for example: if temperature > 50 then do something.
I'm trying to figure out if I can output how many of these devices have met this if > 50 criteria in some time period. The catch is that the rules are defined in real time and should be applied to the Spark job at real time.
How would I do that. Is Spark the right tool for the job?
Many thanks
Is Spark the right tool for the job?
I think so.
the rules are defined in real time and should be applied to the Spark job at real time.
Let's assume the rules are in a database so every batch interval Spark would fetch them and apply one by one. They could also be in a file or any other storage. That's just orthogonal to the main requirement.
How would I do that?
The batch interval would be "some time period". I assume that the payload would have deviceId and temperature. With that you can just use regular filter over temperature and get deviceId back. You don't need stateful pipeline for this unless you want to accumulate data over time that is longer than your batch interval.

How to react on specific event with spark streaming

I'm new to Spark streaming and have following situation:
Multiple (health) devices send their data to my service, every event has at least following data inside (userId, timestamp, pulse, bloodPressure).
In the DB I have per user a threshold for pulse and bloodPressure.
Use Case:
I would like to make a sliding window with Spark streaming which calculates the average per user for pulse and bloodpressure, let's say within 10 min.
After 10 min I would like to check in the DB if the values exceed the threshold per user and execute an action, e.g. call a rest service to send an alarm.
Could somebody tell me if this is generally possible with Spark, and if yes, point me in the right direction?
This is definitely possible. It's not necessarily the best tool to do so though. It depends on the volume of input you expect. If you have hundreds of thousands devices sending one event every second, maybe Spark could be justified. Anyway it's not up to me to validate your architectural choices but keep in mind that resorting to Spark for these use cases make sense only if the volume of data cannot be handled by a single machine.
Also, if the latency of the alert is important and a second or two make a difference, Spark is not the best tool. A processor on a single machine can achieve lower latencies. Otherwise use something more streaming-oriented, like Apache Flink.
As a general advice, if you want to do it in Spark, you just need to create a source (I don't know where your data come from), load the thresholds in a broadcast variable (assuming they are constant over time) and write the logic. To make the rest call, use forEachRdd as the output sink and implement the call logic there.

Perform queries over the time-series stream of data

I'm trying to design an architecture of my streaming application and choose the right tools for the job.
This is how it works currently:
Messages from "application-producer" part have a form of (address_of_sensor, timestamp, content) tuples.
I've already implemented all functionality before Kafka, and now I've encountered major flaw in the design. In "Spark Streaming" part, consolidated stream of messages is translated into stream of events. The problem is that events for the most part are composite - consist of multiple messages, which have occurred at the same time at different sensors.
I can't rely on "time of arrival to Kafka" as a mean to detect "simultaneity". So I has to somehow sort messages in Kafka before extracting them with Spark. Or, more precisely, make queries over Kafka messages.
Maybe Cassandra is the right replacement for Kafka here? I have really simple data model, and only two possible types of queries to perform: query by address, and range query by timestamp. Maybe this is the right choice?
Do somebody have any numbers of Cassandra's throughput?
If you want to run queries on your time series, Cassandra may be the best fit - it is very write optimized, you can build 'wide' rows for your series. It is possible to make slices on your wide rows, so you can select some time ranges with only one query.
On the other hand, kafka can be considered as a raw data flow - you don't have queries, only recently produced data. In order to collect data based on some key in the same partition, you have to select this key carefully. All data within same partition are time sorted.
Range Query on Timestamp is the classic use case of cassandra , if u need address based queries as well u would have to make them as clustering column if using cassandra . As far as cassandra througput are concerned if you can invest in proper performance analysis on cassandra cluster you can achieve very high write throughput . But I have used SparkQL , Cassandra Driver and spark Cassandra connector they don't really give high query throughput speed until you have a big cluster with high CPU configuration , it does not work well with small dataset .
Kafka should not be used as data source for queries , its more of commit log

Resources