Fundamentals of Stream Processing with Apache Beam, Tyler Akidau, Frances Perry

Editor's Notes

• #2: Frances and Tyler talk about Apache Beam -- unified model for batch and stream data processing
• #4: the kind of data set we’re talking about the four questions at the heart of the beam model why these are the right questions the current state of the apache beam project
• #5: talk about the shape of data running example analyzing mobile game logs users around the globe crushing candy want to analyze these logs to learn about game...
• #6: here’s gaming logs each square represents an event where a user scored some points for their team
• #7: game gets popular
• #8: start organizing it into a repeated structure
• #9: repetitive structure just a cheap way of representing an infinite data source. game logs are continuous distributed systems can cause ambiguity...
• #10: Lets look at some points that were scored at 8am <animate> red score 8am, received quickly <animate> yellow score also happened at 8am, received at 8:30 due to network congestion <animate> green element was hours late. this was someone playing in airplane mode on the plane. had to wait for it to land. so now we’ve got an unordered, infinite data set, how do we process it...
• #11: Element wise transformations work on individual elements parsing, translating or filtering applied as elements flow past but other transforms like counting or joining require combining multiple elements together ...
• #12: when doing aggregations, need to divide the infinite stream of elements into finite sized chunks that can be processed independently. simplest way using arrival time in fixed time periods can mean elements are being processed out of order, late elements may be aggregated with unrelated elements that arrived about the same time...
• #13: reorganize data base on when they occurred, not when they arrived red element arrived relatively on time and stays in the noon window. green that arrived at 12:30, was actually created about 11:30, so it moves up to the 11am window. requires formalizing the difference between processing time and event time
• #14: Blue axis is event, Green is processing. Ideally no delay -- elements processed when they occurred <animate> Reality looks more like that red squiggly line, where processing time is slightly delayed off event time. <animate> The variable distance between reality and ideal is called skew. need to track in order to reason about correctness.
• #15: red line the watermark -- no event times earlier than this point are expected to appear in the future. often heuristic based too slow → unnecessary latency. too fast →some data comes in late, after we thought we were done for a given time period. how do we reason about these types of infinite, out-of-order datasets...
• #16: not too hard if you know what kinds of questions to ask!
• #17: What results are calculated? sums, joins, histograms, machine learning models? Where in event time are results calculated? Does the time each event originally occurred affect results? Are results aggregated for all time, in fixed windows, or as user activity sessions? When in processing time are results materialized? Does the time each element arrives in the system affect results? How do we know when to emit a result? What do we do about data that comes in late from those pesky users playing on transatlantic flights? And finally, how do refinements relate? If we choose to emit results multiple times, is each result independent and distinct, do they build upon one another? Let’s dive into how each question contributes when we build a pipeline...
• #18: The first thing to figure out is what you are actually computing. transform each element independently, similar to the Map function in MapReduce, easy to parallelize Other transformations, like Grouping and Combining, require inspecting multiple elements at a time Some operations are really just subgraphs of other more primitive operations. Now let’s see a code snippet for our gaming example...
• #19: Psuedo Java for compactness/clarity! start by reading a collection of raw events transform it into a more structured collection containing key value pairs with a team name and the number of points scored during the event. use a composite operation to sum up all the points per team. Let’s see how this code excutes...
• #20: Looking at points scored for a given team blue axis, green axis <animate> ideal <animate> This score of 3 from just before 12:07 arrives almost immediately. <animate> 7 minutes delayed. elevator or subway. graph not big enough to show offline mode from transatlantic flight
• #21: time is thick white line. accumulate sum into the intermediate state produces output represented by the blue rectangle. all the data available, rectangle covers all events, no matter when in time they occurred. single final result emitted when it’s all complete pretty standard batch processing -> let’s see what happens if we tweak the other questions
• #22: windowing lets us create individual results for different slices of event time. divides data into finite chunks based on the event time of each element. common patterns include fixed time (like hourly, daily, monthly), sliding windows (like the last 24 hours worth of data, every hour) -- single element may be in multiple overlapping windows session based windows that capture bursts of user activity -- unaligned per key very common when trying to aggregations on infinite data also actually common pattern in batch, though historically done using composite keys.
• #23: fixed windows that are 2 minutes long
• #24: independent answer for every two minute period of event time. still waiting until the entire computation completes to emit any results. won’t work for infinite data! want to reduce latency...
• #25: trigger define when in processing time to emit results often relative to the watermark, which is that heuristic about event time progress.
• #26: request that results are emitted when we think we’ve roughly seen all the elements for a given window. actually default -- just written it for clarity.
• #27: left graph shows a perfect watermark -- tracks when all the data for a given event time has arrived emit the result from each window as soon as the watermark passes. <animate> watermark is usually just a heuristic, so look more like graph on the right. now 9 is missed and if the watermark is delayed, like in the first graph, need to wait a long time for anything. would like speculative. lets use a more advanced trigger...
• #28: ask for early, speculative firings every minute get updates every time a late element comes in.
• #29: in all cases, able to get speculative results before the watermark. now get results when watermark passes, but still handle late value 9 even with heuristic watermark in this case, we accumulate across the multiple results per window In the final window, we see and emit 3 but then still include that 3 in the next update of 12. but this behavior around multiple firings is configurable...
• #30: fire three times for a window -- a speculative firing with 3, watermark with two more values 5 and 1, and finally a late value 2. one option is emit new elements that have come in since the last result. requires consumer to be able to do final sum could produce the running sum every time. consumer may overcount produce both the new running sum and retract the old one.
• #31: use accumulating and retracting.
• #32: speculative results, on time results, and retractions. now the final window emits 3, then retracts the 3 when emitting 12. So those are the four questions...
• #33: those are the four key questions are they the right questions?
• #34: here are 5 reasons...
• #35: the results we get are correct this is not something we’ve historically gotten with streaming systems.
• #36: distributed systems are … distributed. if the winds had been blowing from the east instead of the west, elements might have arrived in a slightly different order.
• #37: if we were aggregating based on processing time, this would result in different results for the two orderings.
• #38: aggregating based on event time may have different intermediate results but the final results are identical across the two arrival scenarios.
• #39: next, the abstractions can represent powerful and complex algorithms.
• #40: earlier mentioned session windows -- burst of user activity simple code change...
• #41: want to identify two groupings of points in other words, Tyler was playing the game, got distracted by a squirrel, and then resumed his play.
• #42: now you can see the sessions being built over time at first we see multiple components in the first session not until late element 9 comes in that we realize it’s one big session
• #43: next -- we’ve seen what the four questions can do. what if we ask the questions twice?
• #44: code to calculate the length of a user session
• #45: Remember that these graphs are always shown per key here’s the graph calculating session legths for Frances and the ones for Tyler
• #46: now lets take those session lengths per user ask the questions again this time using fixed windows to take the mean across the entire collection...
• #47: Now calculating the average length of all sessions that ended in a given time period if we rolled out an update to our game, this would let us quickly understand if that resulted a change in user behavior if the change made the game less fun, we could see a sudden drop in how long users play
• #48: ok… flexibility for covering all sorts of uses cases
• #49: By tuning our what/where/when/how knobs, we’ve covered everything from classic batch… to sessions
• #50: And not only that, we do so with lovely modular code
• #51: all these uses cases -- and we never changed our core algorithm just integer summing here, but the same would apply with much more complex algorithms too
• #52: so there you go -- 5 reasons that these 4 questions are awesome
• #53: discussed the use case introduced the four questions in the Beam model shown they are the right questions … now let’s get concrete and see how we got to this model and where we’re going
• #54: Google published the original paper on MapReduce in 2004 -- fundamentally change the way we do distributed processing. <animate> Inside Google, kept innovating, but just published papers <animate> Externally the open source community created Hadoop. Entire ecosystem flourished, partially influenced by those Google papers. <animate> In 2014, Google Cloud Dataflow -- included both a new programming model and fully managed service share this model more broadly -- both because it is awesome and because users benefit from a larger ecosystem and portability across multiple runtimes. So Google, along with a handful of partners donated this programming model to the Apache Software Foundation, as the incubating project Apache Beam...
• #55: Today, Apache Beam includes the core unified programming model the initial Java SDK that we developed as part of Cloud Dataflow and, most important for portability, multiple runners that can execute Beam pipelines on existing distributed process backends. but our longer term goals...
• #56: fully support three different categories of users End users who want to write data processing pipelines Includes adding value like additional connectors -- we’ve got Kafka! Additionally, support community-sourced SDKs and runners Each community has very different sets of goals and needs. having a vision and reaching it are two different things...
• #57: Beam entered incubation in early February. Quickly did the code donations and began bootstrapping the infrastructure. initial focus is on stabilizing internal APIs and integrating the additional runners. Part of that is understanding what different runners can do...
• #58: The Beam model is attempting to generalize semantics -- will not align perfectly with all possible runtimes. Started categorizing the features in the model and the various levels of runner support. This will help users understand mismatches like using event time processing in Spark or exactly once processing with Samza.
• #59: And one of the things we’re most excited about is the collaboration opportunities that Beam enables. Been doing this stuff for a while at Google -- very hermetic environment. Looking forward to incorporating new perspectives -- to build a truly generalizable solution. Growing the Beam development community over the next few months, whether they are looking to write transform libraries for end users, new SDKs, or provide new runners. So if you’re interested in learning more...