Building a Distributed Log from Scratch, Part 3: Scaling Message Delivery

In part two of this series we discussed data replication within the context of a distributed log and how it relates to high availability. Next, we’ll look at what it takes to scale the log such that it can handle non-trivial workloads.

Data Scalability

A key part of scaling any kind of data-intensive system is the ability to partition the data. Partitioning is how we can scale a system linearly, that is to say we can handle more load by adding more nodes. We make the system horizontally scalable.

Kafka was designed this way from the beginning. Topics are partitioned and ordering is only guaranteed within a partition. For example, in an e-commerce application, we might have two topics, purchases and inventory, each with two partitions. These partitions allow us to distribute reads and writes across a set of brokers. In Kafka, the log is actually the partition.

The challenge with this is how we partition the data. We might distribute data using round robin, in effect randomly distributing it. The problem with this is we lose out on ordering, which is an important characteristic of the log. For example, imagine we have add and remove inventory operations. With random partitioning, we might end up with a remove followed by an add getting processed if they’re placed in different partitions. However, if they’re placed in the same partition, we know they will be ordered correctly from the perspective of the publisher.

We could also distribute by hashing a key and sending all writes with the same keys to the same partitions or some custom partitioning strategy along these lines. Continuing with our example, we might partition purchases by account name and inventory by SKU. This way, all purchase operations by the same account are ordered, as are all inventory operations pertaining to the same SKU. The diagram below shows a (naive) custom strategy that partitions topics by ranges based on the account and SKU.

The important point here is that how you partition your data is largely dependent on your application and its usage patterns, but partitioning is a critical part of scalability. It allows you to scale your workload processing by dividing up responsibilities, which in turn, allows you to throw more resources at the problem in a tractable way.

One of NATS Streaming’s shortcomings, in my opinion, is that it doesn’t currently offer a good story around partitioning. Channels are totally ordered, essentially making them the equivalent of a Kafka partition. The workaround is to partition among multiple channels at the application level. To some, this is a benefit because it’s conceptually simpler than Kafka, but Kafka was designed as such because scalability was a key design goal from day one.

Consumer Scalability

One challenge with the log is the problem of high fan-out. Specifically, how do we scale to a large number of consumers? In Kafka and NATS Streaming, reads (and writes) are only served by the leader. Similarly, Amazon Kinesis supports up to only five reads per second per shard (a shard is Kinesis’ equivalent of a partition). Thus, if we have five consumers reading from the same shard, we’ve already hit our fan-out limit. The thought is to partition your workload to increase parallelism and/or daisy chain streams to increase fan-out. But if we are trying to do very high fan-out, e.g. to thousands of IoT devices, neither of these are ideal solutions. Not all use cases may lend themselves to partitioning (though one can argue this is just a sign of poor architecting), and chaining up streams (or in Kafka nomenclature, topics) tends to be kludgey.

However, we can make the following observation: with an immutable log, there are no stale or phantom reads. Unlike a database, we can loosen our requirements a bit. Whereas a database is typically mutable, with a log, we’re only appending things. From a consumer’s perspective, a replica is either up-to-date with the leader or in the process of catching up, but in either case, if we read all of the records, we should end up in the same state. Immutability, at least in theory, should make it “easy” to scale to a large number of consumers because we don’t have to read from the leader to get correct results (ignoring log compaction and other “mutable” operations), so long as we’re okay with strong eventual consistency with respect to tailing the log.

In NATS Streaming, with Raft, we could simply allow followers to serve reads and scale reads by increasing the size of the cluster, but this would impact performance because the quorum size would also increase. Instead, we can use “non-voters” to act as read replicas and balance consumers among them. These read replicas do not participate in quorum or leader election, they simply receive committed log entries. In effect, this is the daisy chaining of streams mentioned earlier but done implicitly by the system. This is an otherwise common pattern for increasing consumer fan-out in Kinesis but is usually done in an ad hoc, Rube Goldberg-esque fashion. Note that, in the case of NATS Streaming, this isn’t quite as simple as it sounds due to the delivery mechanism used, which we’ll describe next.

Push vs. Pull

In Kafka, consumers pull data from brokers. In NATS Streaming, brokers push data to consumers. Kafka’s documentation describes this design decision in detail. The key factor largely comes down to flow control. With push, flow control needs to be explicit to deal with diverse consumers. Different consumers will consume at different rates, so the broker needs to be aware of this so as not to overwhelm a consumer.

There are obvious advantages and disadvantages to both approaches. With push, it can be a tricky balance to ensure full utilization of the consumer. We might use a backoff protocol like additive increase/multiplicative decrease, widely known for its use in TCP congestion control, to optimize utilization. NATS Streaming, like many other messaging systems, implements flow control by using acks. Upon receiving a message, consumers ack back to the server, and the server tracks the in-flight messages for each consumer. If that number goes above a certain threshold, the server will stop delivery until more acks are received. There is a similar flow-control mechanism between the publisher and the server. The trade-off here is the server needs to do some bookkeeping, which we’ll get to in a bit. With a pull-based system, flow control is implicit. Consumers simply go at their own pace, and the server doesn’t need to track anything. There is much less complexity with this.

Pull-based systems lend themselves to aggressive batching. With push, we must decide whether to send a message immediately or wait to accumulate more messages before sending. This is a decision pertaining to latency versus throughput. Push is often viewed as an optimization for latency, but if we’re tuning for low latency, we send messages one at a time only for them to end up being buffered on the consumer anyway. With pull, the consumer fetches all available messages after its current position in the log, which basically removes the guesswork around tuning batching and latency.

There are API implications with this decision too, particularly from an ergonomics and complexity perspective. Kafka clients tend to be “thick” and have a lot of complexity. That is, they do a lot because the broker is designed to be simple. That’s my guess as to why there are so few native client libraries up to par with the Java client. NATS Streaming clients, on the other hand, are relatively “thin” because the server does more. We end up just pushing the complexity around based on our design decisions, but one can argue that the smart client and dumb server is a more scalable approach. We’ll go into detail on that in the next installment of this series.

Circling back on consumer scalability, the fact that NATS Streaming uses a push-based model means we can’t simply setup read replicas and balance consumers among them. Instead, we would need to partition consumers among the replicas so that each server is responsible for pushing data to a subset of consumers. The increased complexity over pull becomes immediately apparent here.

Bookkeeping

There are two ways to track position in the log: have the server track it for consumers or have consumers track it themselves. Again, there are trade-offs with this, namely between API simplicity, server complexity, performance, and scalability. NATS Streaming tracks subscription positions for consumers. This means consumers can come and go as they like and pick back up where they left off easily. Before NATS Streaming supported clustering, this made a lot of sense because the bookkeeping was all in one server. But with clustering, this data must be replicated just like the message data, which poses a performance challenge.

The alternative is to punt the problem to the consumer. But also keep in mind that consumers might not have access to fast stable storage, such as with an IoT device or ephemeral container. Is there a way we can split the difference?

We can store the offsets themselves directly in the log. As of 0.9, this is what Kafka does. Before that, clients had to manage offsets themselves or store them in ZooKeeper. This forced a dependency on ZooKeeper for clients but also posed a major bottleneck since ZooKeeper is relatively low throughput. But by storing offsets in the log, they are treated just like any other write to a Kafka topic, which scales quite well (offsets are stored in an internal Kafka topic called __consumer_offsets partitioned by consumer group; there is also a special read cache for speeding up the read path).

Clients periodically checkpoint their offset to the log. We then use log compaction to retain only the latest offsets. Log compaction works by rewriting the log to retain only the latest message for a given key. On recovery, clients fetch the latest offset from the log. The important part here is we need to structure our keys such that compaction retains the latest offset for each unique consumer. For example, we might structure it as consumer-topic-partition. We end up with something resembling the following, where the message value is the offset:

The above log is uncompacted. Once compacted, it becomes the following:

Note that compaction violates some of our previous assumptions around the immutability of the log, but that’s for another discussion.

There are a number of advantages to this approach. We get fault-tolerance and durability due to the fact that our log is already fault-tolerant and durable as designed earlier. We get consistent reads again due to our replication scheme. Unlike ZooKeeper, we get high write throughput. And we reuse existing structures, so there’s less server complexity. We’re just reusing the log, there aren’t really any major new codepaths.

Interestingly, the bookkeeping needed for flow control in push-based systems—such as acks in NATS Streaming—serves much the same purpose as offset tracking in pull-based systems, since it needs to track position. The difference comes when we allow out-of-order processing. If we don’t allow it, then acks are simply a high-water mark that indicate the client is “this far” caught up. The problem with push is we also have to deal with redeliveries, whereas with pull they are implicitly handled by the client.  If we do allow out-of-order processing, then we need to track individual, in-flight messages, which is what per-message acks allow us to do. In this case, the system starts to look less like a log and more like a message queue. This makes push even more complicated.

The nice thing about reusing the log to track offsets is it greatly reduces the amount of code and complexity needed. Since NATS Streaming allows out-of-order processing, it uses a separate acking subsystem which otherwise has the same requirements as an offset-tracking subsystem.

In part four of this series, we will discuss some of the key trade-offs involved with implementing a distributed log and some lessons learned while building NATS Streaming.

Building a Distributed Log from Scratch, Part 2: Data Replication

In part one of this series we introduced the idea of a message log, touched on why it’s useful, and discussed the storage mechanics behind it. In part two, we discuss data replication.

We have our log. We know how to write data to it and read it back as well as how data is persisted. The caveat to this is, although we have a durable log, it’s a single point of failure (SPOF). If the machine where the log data is stored dies, we’re SOL. Recall that one of our three priorities with this system is high availability, so the question is how do we achieve high availability and fault tolerance?

With high availability, we’re specifically talking about ensuring continuity of reads and writes. A server failing shouldn’t preclude either of these, or at least unavailability should be kept to an absolute minimum and without the need for operator intervention. Ensuring this continuity should be fairly obvious: we eliminate the SPOF. To do that, we replicate the data. Replication can also be a means for increasing scalability, but for now we’re only looking at this through the lens of high availability.

There are a number of ways we can go about replicating the log data. Broadly speaking, we can group the techniques into two different categories: gossip/multicast protocols and consensus protocols. The former includes things like epidemic broadcast trees, bimodal multicast, SWIM, HyParView, and NeEM. These tend to be eventually consistent and/or stochastic. The latter, which I’ve described in more detail here, includes 2PC/3PC, Paxos, Raft, Zab, and chain replication. These tend to favor strong consistency over availability.

So there are a lot of ways we can replicate data, but some of these solutions are better suited than others to this particular problem. Since ordering is an important property of a log, consistency becomes important for a replicated log. If we read from one replica and then read from another, it’s important those views of the log don’t conflict with each other. This more or less rules out the stochastic and eventually consistent options, leaving us with consensus-based replication.

There are essentially two components to consensus-based replication schemes: 1) designate a leader who is responsible for sequencing writes and 2) replicate the writes to the rest of the cluster.

Designating a leader can be as simple as a configuration setting, but the purpose of replication is fault tolerance. If our configured leader crashes, we’re no longer able to accept writes. This means we need the leader to be dynamic. It turns out leader election is a well-understood problem, so we’ll get to this in a bit.

Once a leader is established, it needs to replicate the data to followers. In general, this can be done by either waiting for all replicas or waiting for only a quorum (majority) of replicas. There are pros and cons to both approaches.

Pros Cons
All Replicas Tolerates f failures with f+1 replicas Latency pegged to slowest replica
Quorum Hides delay from a slow replica Tolerates f failures with 2f+1 replicas

Waiting on all replicas means we can make progress as long as at least one replica is available. With quorum, tolerating the same amount of failures requires more replicas because we need a majority to make progress. The trade-off is that the quorum hides any delays from a slow replica. Kafka is an example of a system which uses all replicas (with some conditions on this which we will see later), and NATS Streaming is one that uses a quorum. Let’s take a look at both in more detail.

Replication in Kafka

In Kafka, a leader is selected (we’ll touch on this in a moment). This leader maintains an in-sync replica set (ISR) consisting of all the replicas which are fully caught up with the leader. This is every replica, by definition, at the beginning. All reads and writes go through the leader. The leader writes messages to a write-ahead log (WAL). Messages written to the WAL are considered uncommitted or “dirty” initially. The leader only commits a message once all replicas in the ISR have written it to their own WAL. The leader also maintains a high-water mark (HW) which is the last committed message in the WAL. This gets piggybacked on the replica fetch responses from which replicas periodically checkpoint to disk for recovery purposes. The piggybacked HW then allows replicas to know when to commit.

Only committed messages are exposed to consumers. However, producers can configure how they want to receive acknowledgements on writes. It can wait until the message is committed on the leader (and thus replicated to the ISR), wait for the message to only be written (but not committed) to the leader’s WAL, or not wait at all. This all depends on what trade-offs the producer wants to make between latency and durability.

The graphic below shows how this replication process works for a cluster of three brokers: b1, b2, and b3. Followers are effectively special consumers of the leader’s log.

Now let’s look at a few failure modes and how Kafka handles them.

Leader Fails

Kafka relies on Apache ZooKeeper for certain cluster coordination tasks, such as leader election, though this is not actually how the log leader is elected. A Kafka cluster has a single controller broker whose election is handled by ZooKeeper. This controller is responsible for performing administrative tasks on the cluster. One of these tasks is selecting a new log leader (actually partition leader, but this will be described later in the series) from the ISR when the current leader dies. ZooKeeper is also used to detect these broker failures and signal them to the controller.

Thus, when the leader crashes, the cluster controller is notified by ZooKeeper and it selects a new leader from the ISR and announces this to the followers. This gives us automatic failover of the leader. All committed messages up to the HW are preserved and uncommitted messages may be lost during the failover. In this case, b1 fails and b2 steps up as leader.

Follower Fails

The leader tracks information on how “caught up” each replica is. Before Kafka 0.9, this included both how many messages a replica was behind, replica.lag.max.messages, and the amount of time since the replica last fetched messages from the leader, replica.lag.time.max.ms. Since 0.9, replica.lag.max.messages was removed and replica.lag.time.max.ms now refers to both the time since the last fetch request and the amount of time since the replica last caught up.

Thus, when a follower fails (or stops fetching messages for whatever reason), the leader will detect this based on replica.lag.time.max.ms. After that time expires, the leader will consider the replica out of sync and remove it from the ISR. In this scenario, the cluster enters an “under-replicated” state since the ISR has shrunk. Specifically, b2 fails and is removed from the ISR.

Follower Temporarily Partitioned

The case of a follower being temporarily partitioned, e.g. due to a transient network failure, is handled in a similar fashion to the follower itself failing. These two failure modes can really be combined since the latter is just the former with an arbitrarily long partition, i.e. it’s the difference between crash-stop and crash-recovery models.

In this case, b3 is partitioned from the leader. As before, replica.lag.time.max.ms acts as our failure detector and causes b3 to be removed from the ISR. We enter an under-replicated state and the remaining two brokers continue committing messages 4 and 5. Accordingly, the HW is updated to 5 on these brokers.

When the partition heals, b3 continues reading from the leader and catching up. Once it is fully caught up with the leader, it’s added back into the ISR and the cluster resumes its fully replicated state.

We can generalize this to the crash-recovery model. For example, instead of a network partition, the follower could crash and be restarted later. When the failed replica is restarted, it recovers the HW from disk and truncates its log up to the HW. This preserves the invariant that messages after the HW are not guaranteed to be committed. At this point, it can begin catching up from the leader and will end up with a log consistent with the leader’s once fully caught up.

Replication in NATS Streaming

NATS Streaming relies on the Raft consensus algorithm for leader election and data replication. This sometimes comes as a surprise to some as Raft is largely seen as a protocol for replicated state machines. We’ll try to understand why Raft was chosen for this particular problem in the following sections. We won’t dive deep into Raft itself beyond what is needed for the purposes of this discussion.

While a log is a state machine, it’s a very simple one: a series of appends. Raft is frequently used as the replication mechanism for key-value stores which have a clearer notion of “state machine.” For example, with a key-value store, we have set and delete operations. If we set foo = bar and then later set foo = baz, the state gets rolled up. That is, we don’t necessarily care about the provenance of the key, only its current state.

However, NATS Streaming differs from Kafka in a number of key ways. One of these differences is that NATS Streaming attempts to provide a sort of unified API for streaming and queueing semantics not too dissimilar from Apache Pulsar. This means, while it has a notion of a log, it also has subscriptions on that log. Unlike Kafka, NATS Streaming tracks these subscriptions and metadata associated with them, such as where a client is in the log. These have definite “state machines” affiliated with them, like creating and deleting subscriptions, positions in the log, clients joining or leaving queue groups, and message-redelivery information.

Currently, NATS Streaming uses multiple Raft groups for replication. There is a single metadata Raft group used for replicating client state and there is a separate Raft group per topic which replicates messages and subscriptions.

Raft solves both the problems of leader election and data replication in a single protocol. The Secret Lives of Data provides an excellent interactive illustration of how this works. As you step through that illustration, you’ll notice that the algorithm is actually quite similar to the Kafka replication protocol we walked through earlier. This is because although Raft is used to implement replicated state machines, it actually is a replicated WAL, which is exactly what Kafka is. One benefit of using Raft is we no longer have the need for ZooKeeper or some other coordination service.

Raft handles electing a leader. Heartbeats are used to maintain leadership. Writes flow through the leader to the followers. The leader appends writes to its WAL and they are subsequently piggybacked onto the heartbeats which get sent to the followers using AppendEntries messages. At this point, the followers append the write to their own WALs, assuming they don’t detect a gap, and send a response back to the leader. The leader commits the write once it receives a successful response from a quorum of followers.

Similar to Kafka, each replica in Raft maintains a high-water mark of sorts called the commit index, which is the index of the highest log entry known to be committed. This is piggybacked on the AppendEntries messages which the followers use to know when to commit entries in their WALs. If a follower detects that it missed an entry (i.e. there was a gap in the log), it rejects the AppendEntries and informs the leader to rewind the replication. The Raft paper details how it ensures correctness, even in the face of many failure modes such as the ones described earlier.

Conceptually, there are two logs: the Raft log and the NATS Streaming message log. The Raft log handles replicating messages and, once committed, they are appended to the NATS Streaming log. If it seems like there’s some redundancy here, that’s because there is, which we’ll get to soon. However, keep in mind we’re not just replicating the message log, but also the state machines associated with the log and any clients.

There are a few challenges with this replication technique, two of which we will talk about. The first is scaling Raft. With a single topic, there is one Raft group, which means one node is elected leader and it heartbeats messages to followers.

As the number of topics increases, so do the number of Raft groups, each with their own leaders and heartbeats. Unless we constrain the Raft group participants or the number of topics, this creates an explosion of network traffic between nodes.

There are a couple ways we can go about addressing this. One option is to run a fixed number of Raft groups and use a consistent hash to map a topic to a group. This can work well if we know roughly the number of topics beforehand since we can size the number of Raft groups accordingly. If you expect only 10 topics, running 10 Raft groups is probably reasonable. But if you expect 10,000 topics, you probably don’t want 10,000 Raft groups. If hashing is consistent, it would be feasible to dynamically add or remove Raft groups at runtime, but it would still require repartitioning a portion of topics which can be complicated.

Another option is to run an entire node’s worth of topics as a single group using a layer on top of Raft. This is what CockroachDB does to scale Raft in proportion to the number of key ranges using a layer on top of Raft they call MultiRaft. This requires some cooperation from the Raft implementation, so it’s a bit more involved than the partitioning technique but eschews the repartitioning problem and redundant heartbeating.

The second challenge with using Raft for this problem is the issue of “dual writes.” As mentioned before, there are really two logs: the Raft log and the NATS Streaming message log, which we’ll call the “store.” When a message is published, the leader writes it to its Raft log and it goes through the Raft replication process.

Once the message is committed in Raft, it’s written to the NATS Streaming log and the message is now visible to consumers.

Note, however, that not only messages are written to the Raft log. We also have subscriptions and cluster topology changes, for instance. These other items are not written to the NATS Streaming log but handled in other ways on commit. That said, messages tend to occur in much greater volume than these other entries.

Messages end up getting stored redundantly, once in the Raft log and once in the NATS Streaming log. We can address this problem if we think about our logs a bit differently. If you recall from part one, our log storage consists of two parts: the log segment and the log index. The segment stores the actual log data, and the index stores a mapping from log offset to position in the segment.

Along these lines, we can think of the Raft log index as a “physical offset” and the NATS Streaming log index as a “logical offset.” Instead of maintaining two logs, we treat the Raft log as our message write-ahead log and treat the NATS Streaming log as an index into that WAL. Particularly, messages are written to the Raft log as usual. Once committed, we write an index entry for the message offset that points back into the log. As before, we use the index to do lookups into the log and can then read sequentially from the log itself.

Remaining Questions

We’ve answered the questions of how to ensure continuity of reads and writes, how to replicate data, and how to ensure replicas are consistent. The remaining two questions pertaining to replication are how do we keep things fast and how do we ensure data is durable?

There are several things we can do with respect to performance. The first is we can configure publisher acks depending on our application’s requirements. Specifically, we have three options. The first is the broker acks on commit. This is slow but safe as it guarantees the data is replicated. The second is the broker acks on appending to its local log. This is fast but unsafe since it doesn’t wait on any replica roundtrips but, by that very fact, means that the data is not replicated. If the leader crashes, the message could be lost. Lastly, the publisher can just not wait for an ack at all. This is the fastest but least safe option for obvious reasons. Tuning this all depends on what requirements and trade-offs make sense for your application.

The second thing we do is don’t explicitly fsync writes on the broker and instead rely on replication for durability. Both Kafka and NATS Streaming (when clustered) do this. With fsync enabled (in Kafka, this is configured with flush.messages and/or flush.ms and in NATS Streaming, with file_sync), every message that gets published results in a sync to disk. This ends up being very expensive. The thought here is if we are replicating to enough nodes, the replication itself is sufficient for HA of data since the likelihood of more than a quorum of nodes failing is low, especially if we are using rack-aware clustering. Note that data is still periodically flushed in the background by the kernel.

Batching aggressively is also a key part of ensuring good performance. Kafka supports end-to-end batching from the producer all the way to the consumer. NATS Streaming does not currently support batching at the API level, but it uses aggressive batching when replicating and persisting messages. In my experience, this makes about an order-of-magnitude improvement in throughput.

Finally, as already discussed earlier in the series, keeping disk access sequential and maximizing zero-copy reads makes a big difference as well.

There are a few things worth noting with respect to durability. Quorum is what guarantees durability of data. This comes “for free” with Raft due to the nature of that protocol. In Kafka, we need to do a bit of configuring to ensure this. Namely, we need to configure min.insync.replicas on the broker and acks on the producer. The former controls the minimum number of replicas that must acknowledge a write for it to be considered successful when a producer sets acks to “all.” The latter controls the number of acknowledgments the producer requires the leader to have received before considering a request complete. For example, with a topic that has a replication factor of three, min.insync.replicas needs to be set to two and acks set to “all.” This will, in effect, require a quorum of two replicas to process writes.

Another caveat with Kafka is unclean leader elections. That is, if all replicas become unavailable, there are two options: choose the first replica to come back to life (not necessarily in the ISR) and elect this replica as leader (which could result in data loss) or wait for a replica in the ISR to come back to life and elect it as leader (which could result in prolonged unavailability). By default, Kafka favors availability by choosing the second strategy. If you prefer consistency, you must set unclean.leader.election.enable to false.

Fundamentally, durability and consistency are at odds with availability. If there is no quorum, then no reads or writes can be accepted and the cluster is unavailable. This is the crux of the CAP theorem.

In part three of this series, we will discuss scaling message delivery in the distributed log.

Benchmarking Commit Logs

In this article, we look at Apache Kafka and NATS Streaming, two messaging systems based on the idea of a commit log. We’ll compare some of the features of both but spend less time talking about Kafka since by now it’s quite well known. Similar to previous studies, we’ll attempt to quantify their general performance characteristics through careful benchmarking.

The purpose of this benchmark is to test drive the newly released NATS Streaming system, which was made generally available just in the last few months. NATS Streaming doesn’t yet support clustering, so we try to put its performance into context by looking at a similar configuration of Kafka.

Unlike conventional message queues, commit logs are an append-only data structure. This results in several nice properties like total ordering of messages, at-least-once delivery, and message-replay semantics. Jay Kreps’ blog post The Log is a great introduction to the concept and particularly why it’s so useful in the context of distributed systems and stream processing (his book I Heart Logs is an extended version of the blog post and is a quick read).

Kafka, which originated at LinkedIn, is by far the most popular and most mature implementation of the commit log (AWS offers their own flavor of it called Kinesis, and imitation is the sincerest form of flattery). It’s billed as a “distributed streaming platform for building real-time data pipelines and streaming apps.” The much newer NATS Streaming is actually a data-streaming layer built on top of Apcera’s high-performance publish-subscribe system NATS. It’s billed as “real-time streaming for Big Data, IoT, Mobile, and Cloud Native Applications.” Both have some similarities as well as some key differences.

Fundamental to the notion of a log is a way to globally order events. Neither NATS Streaming nor Kafka are actually a single log but many logs, each totally ordered using a sequence number or offset, respectively.

In Kafka, topics are partitioned into multiple logs which are then replicated across a number of servers for fault tolerance, making it a distributed commit log. Each partition has a server that acts as the leader. Cluster membership and leader election is managed by ZooKeeper.

NATS Streaming’s topics are called “channels” which are globally ordered. Unlike Kafka, NATS Streaming does not support replication or partitioning of channels, though my understanding is clustering support is slated for Q1 2017. Its message store is pluggable, so it can provide durability using a file-backed implementation, like Kafka, or simply an in-memory store.

NATS Streaming is closer to a hybrid of traditional message queues and the commit log. Like Kafka, it allows replaying the log from a specific offset, the beginning of time, or the newest offset, but it also exposes an API for reading from the log at a specific physical time offset, e.g. all messages from the last 30 seconds. Kafka, on the other hand, only has a notion of logical offsets (correction: Kafka added support for offset lookup by timestamp in 0.10.1.0) . Generally, relying on physical time is an anti-pattern in distributed systems due to clock drift and the fact that clocks are not always monotonic. For example, imagine a situation where a NATS Streaming server is restarted and the clock is changed. Messages are still ordered by their sequence numbers but their timestamps might not reflect that. Developers would need to be aware of this while implementing their business logic.

With Kafka, it’s strictly on consumers to track their offset into the log (or the high-level consumer which stores offsets in ZooKeeper (correction: Kafka itself can now store offsets which is used by the new Consumer API, meaning clients do not have to manage offsets directly or rely on ZooKeeper)). NATS Streaming allows clients to either track their sequence number or use a durable subscription, which causes the server to track the last acknowledged message for a client. If the client restarts, the server will resume delivery starting at the earliest unacknowledged message. This is closer to what you would expect from a traditional message-oriented middleware like RabbitMQ.

Lastly, NATS Streaming supports publisher and subscriber rate limiting. This works by configuring the maximum number of in-flight (unacknowledged) messages either from the publisher to the server or from the server to the subscriber. Starting in version 0.9, Kafka supports a similar rate limiting feature that allows producer and consumer byte-rate thresholds to be defined for groups of clients with its Quotas protocol.

Kafka was designed to avoid tracking any client state on the server for performance and scalability reasons. Throughput and storage capacity scale linearly with the number of nodes. NATS Streaming provides some additional features over Kafka at the cost of some added state on the server. Since clustering isn’t supported, there isn’t really any scale or HA story yet, so it’s unclear how that will play out. That said, once replication is supported, there’s a lot of work going into verifying its correctness (which is a major advantage Kafka has).

Benchmarks

Since NATS Streaming does not support replication at this time (0.3.1), we’ll compare running a single instance of it with file-backed persistence to running a single instance of Kafka (0.10.1.0). We’ll look at both latency and throughput running on commodity hardware (m4.xlarge EC2 instances) with load generation and consumption each running on separate instances. In all of these benchmarks, the systems under test have not been tuned at all and are essentially in their “off-the-shelf” configurations.

We’ll first look at latency by publishing messages of various sizes, ranging from 256 bytes to 1MB, at a fixed rate of 50 messages/second for 30 seconds. Message contents are randomized to account for compression. We then plot the latency distribution by percentile on a logarithmic scale from the 0th percentile to the 99.9999th percentile. Benchmarks are run several times in an attempt to produce a “normalized” result. The benchmark code used is open source.

First, to establish a baseline and later get a feel for the overhead added by the file system, we’ll benchmark NATS Streaming with in-memory storage, meaning messages are not written to disk.

Unsurprisingly, the 1MB configuration has much higher latencies than the other configurations, but everything falls within single-digit-millisecond latencies.nats_mem

NATS Streaming 0.3.1 (in-memory persistence)

 Size 99% 99.9% 99.99% 99.999% 99.9999% 
256B 0.3750ms 1.0367ms 1.1257ms 1.1257ms 1.1257ms
1KB 0.38064ms 0.8321ms 1.3260ms 1.3260ms 1.3260ms
5KB 0.4408ms 1.7569ms 2.1465ms 2.1465ms 2.1465ms
1MB 6.6337ms 8.8097ms 9.5263ms 9.5263ms 9.5263ms

Next, we look at NATS Streaming with file-backed persistence. This provides the same durability guarantees as Kafka running with a replication factor of 1. By default, Kafka stores logs under /tmp. Many Unix distributions mount /tmp to tmpfs which appears as a mounted file system but is actually stored in volatile memory. To account for this and provide as level a playing field as possible, we configure NATS Streaming to also store its logs in /tmp.

As expected, latencies increase by about an order of magnitude once we start going to disk.

nats_file_fsync

NATS Streaming 0.3.1 (file-backed persistence)

 Size 99% 99.9% 99.99% 99.999% 99.9999% 
256B 21.7051ms 25.0369ms 27.0524ms 27.0524ms 27.0524ms
1KB 20.6090ms 23.8858ms 24.7124ms 24.7124ms 24.7124ms
5KB 22.1692ms 35.7394ms 40.5612ms 40.5612ms 40.5612ms
1MB 45.2490ms 130.3972ms 141.1564ms 141.1564ms 141.1564ms

Since we will be looking at Kafka, there is an important thing to consider relating to fsync behavior. As of version 0.8, Kafka does not call fsync directly and instead relies entirely on the background flush performed by the OS. This is clearly indicated by their documentation:

We recommend using the default flush settings which disable application fsync entirely. This means relying on the background flush done by the OS and Kafka’s own background flush. This provides the best of all worlds for most uses: no knobs to tune, great throughput and latency, and full recovery guarantees. We generally feel that the guarantees provided by replication are stronger than sync to local disk, however the paranoid still may prefer having both and application level fsync policies are still supported.

However, NATS Streaming calls fsync every time a batch is written to disk by default. This can be disabled through the use of the –file_sync flag. By setting this flag to false, we put NATS Streaming’s persistence behavior closer in line with Kafka’s (again assuming a replication factor of 1).

As an aside, the comparison between NATS Streaming and Kafka still isn’t completely “fair”. Jay Kreps points out that Kafka relies on replication as the primary means of durability.

Kafka leaves [fsync] off by default because it relies on replication not fsync for durability, which is generally faster. If you don’t have replication I think you probably need fsync and maybe some kind of high integrity file system.

I don’t think we can provide a truly fair comparison until NATS Streaming supports replication, at which point we will revisit this.

To no one’s surprise, setting –file_sync=false has a significant impact on latency, shown in the distribution below.

nats_file_no_fsync

In fact, it’s now in line with the in-memory performance as before for 256B, 1KB, and 5KB messages, shown in the comparison below.

nats_file_mem

For a reason I have yet to figure out, the latency for 1MB messages is roughly an order of magnitude faster when fsync is enabled after the 95th percentile, which seems counterintuitive. If anyone has an explanation, I would love to hear it. I’m sure there’s a good debug story there. The distribution below shows the 1MB configuration for NATS Streaming with and without fsync enabled and just how big the difference is at the 95th percentile and beyond.

nats_file_mem_1mb

NATS Streaming 0.3.1 (file-backed persistence, –file_sync=false)

 Size 99% 99.9% 99.99% 99.999% 99.9999% 
256B 0.4304ms 0.8577ms 1.0706ms 1.0706ms 1.0706ms
1KB 0.4372ms 1.5987ms 1.8651ms 1.8651ms 1.8651ms
5KB 0.4939ms 2.0828ms 2.2540ms 2.2540ms 2.2540ms
1MB 1296.1464ms 1556.1441ms 1596.1457ms 1596.1457ms 1596.1457ms

Kafka with replication factor 1 tends to have higher latencies than NATS Streaming with –file_sync=false. There was one potential caveat here Ivan Kozlovic pointed out to me in that NATS Streaming uses a caching optimization for reads that may put it at an advantage.

Now, there is one side where NATS Streaming *may* be looking better and not fair to Kafka. By default, the file store keeps everything in memory once stored. This means look-ups will be fast. There is only a all-or-nothing mode right now, which means either cache everything or nothing. With caching disabled (–file_cache=false), every lookup will result in disk access (which when you have 1 to many subscribers will be bad). I am working on changing that. But if you do notice that in Kafka, consuming results in a disk read (given the other default behavior described above, they actually may not ;-)., then you could disable NATS Streaming file caching.

Fortunately, we can verify if Kafka is actually going to disk to read messages back from the log during the benchmark using iostat. We see something like this for the majority of the benchmark duration:

avg-cpu:  %user   %nice %system %iowait  %steal   %idle
          13.53    0.00   11.28    0.00    0.00   75.19

Device:    tps   Blk_read/s   Blk_wrtn/s   Blk_read   Blk_wrtn
xvda      0.00         0.00         0.00          0          0

Specifically, we’re interested in Blk_read, which indicates the total number of blocks read. It appears that Kafka does indeed make heavy use of the operating system’s page cache as Blk_wrtn and Blk_read rarely show any activity throughout the entire benchmark. As such, it seems fair to leave NATS Streaming’s –file_cache=true, which is the default.

One interesting point is Kafka offloads much of its caching to the page cache and outside of the JVM heap, clearly in an effort to minimize GC pauses. I’m not clear if the cache Ivan refers to in NATS Streaming is off-heap or not (NATS Streaming is written in Go which, like Java, is a garbage-collected language).

Below is the distribution of latencies for 256B, 1KB, and 5KB configurations in Kafka.

kafka

Similar to NATS Streaming, 1MB message latencies tend to be orders of magnitude worse after about the 80th percentile. The distribution below compares the 1MB configuration for NATS Streaming and Kafka.

nats_kafka_1mb

Kafka 0.10.1.0 (replication factor 1)

 Size 99% 99.9% 99.99% 99.999% 99.9999% 
256B 0.9230ms 1.4575ms 1.6596ms 1.6596ms 1.6596ms
1KB 0.5942ms 1.3123ms 17.6556ms 17.6556ms 17.6556ms
5KB 0.7203ms 5.7236ms 18.9334ms 18.9334ms 18.9334ms
1MB 5337.3174ms 5597.3315ms 5617.3199ms 5617.3199ms 5617.3199ms

The percentile distributions below compare NATS Streaming and Kafka for the 256B, 1KB, and 5KB configurations, respectively.

nats_kafka_256b

nats_kafka_1kb

nats_kafka_5kb

Next, we’ll look at overall throughput for the two systems. This is done by publishing 100,000 messages using the same range of sizes as before and measuring the elapsed time. Specifically, we measure throughput at the publisher and the subscriber.

Despite using an asynchronous publisher in both the NATS Streaming and Kafka benchmarks, we do not consider the publisher “complete” until it has received acks for all published messages from the server. In Kafka, we do this by setting request.required.acks to 1, which means the leader replica has received the data, and consuming the received acks. This is important because the default value is 0, which means the producer never waits for an ack from the broker. In NATS Streaming, we provide an ack callback on every publish. We use the same benchmark configuration as the latency tests, separating load generation and consumption on different EC2 instances. Note the log scale in the following charts.

Once again, we’ll start by looking at NATS Streaming using in-memory persistence. The truncated 1MB send and receive throughputs are 93.01 messages/second.

nats_mem_throughput

For comparison, we now look at NATS Streaming with file persistence and –file_sync=false. As before, this provides the closest behavior to Kafka’s default flush behavior. The second chart shows a side-by-side comparison between NATS Streaming with in-memory and file persistence.

nats_file_throughput

nats_compare_throughput

Lastly, we look at Kafka with replication factor 1. Throughput significantly deteriorates when we set request.required.acks = 1 since the producer must wait for all acks from the server. This is important though because, by default, the client does not require an ack from the server. If this were the case, the producer would have no idea how much data actually reached the server once it finished—it could simply be buffered in the client, in flight over the wire, or in the server but not yet on disk. Running the benchmark with request.required.acks = 0 yields much higher throughput on the sender but is basically an exercise in how fast you can write to a channel using the Sarama Go client—slightly misleading.

kafka_throughput

Looking at some comparisons of Kafka and NATS Streaming, we can see that NATS Streaming has higher throughput in all but a few cases.

nats_kafka_throughput

nats_kafka_send_throughput

I want to repeat the disclaimer from before: the purpose of this benchmark is to test drive the newly released NATS Streaming system (which as mentioned earlier, doesn’t yet support clustering), and put its performance into context by looking at a similar configuration of Kafka.

Kafka generally scales very well, so measuring the throughput of a single broker with a single producer and single consumer isn’t particularly meaningful. In reality, we’d be running a cluster with several brokers and partitioning our topics across them.

For as young as it is, NATS Streaming has solid performance (which shouldn’t come as much of a surprise considering the history of NATS itself), and I imagine it will only get better with time as the NATS team continues to optimize. In some ways, NATS Streaming bridges the gap between the commit log as made popular by Kafka and the conventional message queue as made popular by protocols like JMS, AMQP, STOMP, and the like.

The bigger question at this point is how NATS Streaming will tackle scaling and replication (a requirement for true production-readiness in my opinion). Kafka was designed from the ground up for high scalability and availability through the use of external coordination (read ZooKeeper). Naturally, there is a lot of complexity and cost that comes with that. NATS Streaming attempts to keep NATS’ spirit of simplicity, but it’s yet to be seen how it will reconcile that with the complex nature of distributed systems. I’m excited to see where Apcera takes NATS Streaming and generally the NATS ecosystem in the future since the team has a lot of experience in this area.

You Cannot Have Exactly-Once Delivery

I’m often surprised that people continually have fundamental misconceptions about how distributed systems behave. I myself shared many of these misconceptions, so I try not to demean or dismiss but rather educate and enlighten, hopefully while sounding less preachy than that just did. I continue to learn only by following in the footsteps of others. In retrospect, it shouldn’t be surprising that folks buy into these fallacies as I once did, but it can be frustrating when trying to communicate certain design decisions and constraints.

Within the context of a distributed system, you cannot have exactly-once message delivery. Web browser and server? Distributed. Server and database? Distributed. Server and message queue? Distributed. You cannot have exactly-once delivery semantics in any of these situations.

As I’ve described in the past, distributed systems are all about trade-offs. This is one of them. There are essentially three types of delivery semantics: at-most-once, at-least-once, and exactly-once. Of the three, the first two are feasible and widely used. If you want to be super anal, you might say at-least-once delivery is also impossible because, technically speaking, network partitions are not strictly time-bound. If the connection from you to the server is interrupted indefinitely, you can’t deliver anything. Practically speaking, you have bigger fish to fry at that point—like calling your ISP—so we consider at-least-once delivery, for all intents and purposes, possible. With this model of thinking, network partitions are finitely bounded in time, however arbitrary this may be.

So where does the trade-off come into play, and why is exactly-once delivery impossible? The answer lies in the Two Generals thought experiment or the more generalized Byzantine Generals Problem, which I’ve looked at extensively. We must also consider the FLP result, which basically says, given the possibility of a faulty process, it’s impossible for a system of processes to agree on a decision.

In the letter I mail you, I ask you to call me once you receive it. You never do. Either you really didn’t care for my letter or it got lost in the mail. That’s the cost of doing business. I can send the one letter and hope you get it, or I can send 10 letters and assume you’ll get at least one of them. The trade-off here is quite clear (postage is expensive!), but sending 10 letters doesn’t really provide any additional guarantees. In a distributed system, we try to guarantee the delivery of a message by waiting for an acknowledgement that it was received, but all sorts of things can go wrong. Did the message get dropped? Did the ack get dropped? Did the receiver crash? Are they just slow? Is the network slow? Am slow? FLP and the Two Generals Problem are not design complexities, they are impossibility results.

People often bend the meaning of “delivery” in order to make their system fit the semantics of exactly-once, or in other cases, the term is overloaded to mean something entirely different. State-machine replication is a good example of this. Atomic broadcast protocols ensure messages are delivered reliably and in order. The truth is, we can’t deliver messages reliably and in order in the face of network partitions and crashes without a high degree of coordination. This coordination, of course, comes at a cost (latency and availability), while still relying on at-least-once semantics. Zab, the atomic broadcast protocol which lays the foundation for ZooKeeper, enforces idempotent operations.

State changes are idempotent and applying the same state change multiple times does not lead to inconsistencies as long as the application order is consistent with the delivery order. Consequently, guaranteeing at-least once semantics is sufficient and simplifies the implementation.

“Simplifies the implementation” is the authors’ attempt at subtlety. State-machine replication is just that, replicating state. If our messages have side effects, all of this goes out the window.

We’re left with a few options, all equally tenuous. When a message is delivered, it’s acknowledged immediately before processing. The sender receives the ack and calls it a day. However, if the receiver crashes before or during its processing, that data is lost forever. Customer transaction? Sorry, looks like you’re not getting your order. This is the worldview of at-most-once delivery. To be honest, implementing at-most-once semantics is more complicated than this depending on the situation. If there are multiple workers processing tasks or the work queues are replicated, the broker must be strongly consistent (or CP in CAP theorem parlance) so as to ensure a task is not delivered to any other workers once it’s been acked. Apache Kafka uses ZooKeeper to handle this coordination.

On the other hand, we can acknowledge messages after they are processed. If the process crashes after handling a message but before acking (or the ack isn’t delivered), the sender will redeliver. Hello, at-least-once delivery. Furthermore, if you want to deliver messages in order to more than one site, you need an atomic broadcast which is a huge burden on throughput. Fast or consistent. Welcome to the world of distributed systems.

Every major message queue in existence which provides any guarantees will market itself as at-least-once delivery. If it claims exactly-once, it’s because they are lying to your face in hopes that you will buy it or they themselves do not understand distributed systems. Either way, it’s not a good indicator.

RabbitMQ attempts to provide guarantees along these lines:

When using confirms, producers recovering from a channel or connection failure should retransmit any messages for which an acknowledgement has not been received from the broker. There is a possibility of message duplication here, because the broker might have sent a confirmation that never reached the producer (due to network failures, etc). Therefore consumer applications will need to perform deduplication or handle incoming messages in an idempotent manner.

The way we achieve exactly-once delivery in practice is by faking it. Either the messages themselves should be idempotent, meaning they can be applied more than once without adverse effects, or we remove the need for idempotency through deduplication. Ideally, our messages don’t require strict ordering and are commutative instead. There are design implications and trade-offs involved with whichever route you take, but this is the reality in which we must live.

Rethinking operations as idempotent actions might be easier said than done, but it mostly requires a change in the way we think about state. This is best described by revisiting the replicated state machine. Rather than distributing operations to apply at various nodes, what if we just distribute the state changes themselves? Rather than mutating state, let’s just report facts at various points in time. This is effectively how Zab works.

Imagine we want to tell a friend to come pick us up. We send him a series of text messages with turn-by-turn directions, but one of the messages is delivered twice! Our friend isn’t too happy when he finds himself in the bad part of town. Instead, let’s just tell him where we are and let him figure it out. If the message gets delivered more than once, it won’t matter. The implications are wider reaching than this, since we’re still concerned with the ordering of messages, which is why solutions like commutative and convergent replicated data types are becoming more popular. That said, we can typically solve this problem through extrinsic means like sequencing, vector clocks, or other partial-ordering mechanisms. It’s usually causal ordering that we’re after anyway. People who say otherwise don’t quite realize that there is no now in a distributed system.

To reiterate, there is no such thing as exactly-once delivery. We must choose between the lesser of two evils, which is at-least-once delivery in most cases. This can be used to simulate exactly-once semantics by ensuring idempotency or otherwise eliminating side effects from operations. Once again, it’s important to understand the trade-offs involved when designing distributed systems. There is asynchrony abound, which means you cannot expect synchronous, guaranteed behavior. Design for failure and resiliency against this asynchronous nature.

Not Invented Here

Engineers love engineering things. The reason is self-evident (and maybe self-fulfilling—why else would you be an engineer?). We like to think we’re pretty good at solving problems. Unfortunately, this mindset can, on occasion, yield undesirable consequences which might not be immediately apparent but all the while damaging.

Developers are all in tune with the idea of “don’t reinvent the wheel,” but it seems to be eschewed sometimes, deliberately or otherwise. People don’t generally write their own merge sort, so why would they write their own consensus protocol? Anecdotally speaking, they do.

Not-Invented-Here Syndrome is a very real thing. In many cases, consciously or not, it’s a cultural problem. In others, it’s an engineering one. Camille Fournier’s blog post on ZooKeeper helps to illustrate this point and provide some context. In it, she describes why some distributed systems choose to rely on external services, such as ZooKeeper, for distributed coordination, while others build in their own coordination logic.

We draw a parallel between distributed systems and traditional RDBMSs, which typically implement their own file system and other low-level facilities. Why? Because it’s their competitive advantage. SQL databases sell because they offer finely tuned performance, and in order to do that, they need to control these things that the OS otherwise provides. Distributed databases like Riak sell because they own the coordination logic, which helps promote their competitive advantage. This follows what Joel Spolsky says about NIH Syndrome in that “if it’s a core business function—do it yourself, no matter what.”

If you’re developing a computer game where the plot is your competitive advantage, it’s OK to use a third party 3D library. But if cool 3D effects are going to be your distinguishing feature, you had better roll your own.

This makes a lot of sense. My sorting algorithm is unlikely to provide me with a competitive edge, but something else might, even if it’s not particularly novel.

So in some situations, homegrown is justifiable, but that’s not always the case. Redis’ competitive advantage is its predictably low latencies and data structures. Does it make sense for it to implement its own clustering and leader election protocols? Maybe, but this is where NIH can bite you. If what you’re doing is important and there’s precedent, lean on existing research and solutions. Most would argue write safety is important, and there is certainly precedent for leader election. Why not leverage that work? Things like Raft, Paxos, and Zab provide solutions which are proven using formal methods and are peer reviewed. That doesn’t mean new solutions can’t be developed, but they generally require model checking and further scrutiny to ensure correctness. Otherwise, you’ll inevitably run into problems. Implementing our own solutions can provide valuable insight, but leave them at home if they’re not rigorously approached. Rolling your own and calling it “good enough” is dishonest to your users if it’s not properly communicated.

Elasticsearch is another interesting case to look at. You might say Elasticsearch’s competitive advantage is its full-text search engine, but it’s not. Like Solr, it’s built on Lucene. Elasticsearch was designed from the ground-up to be distributed. This is what gives it a leg up over Solr and other similar search servers where horizontal scaling and fault tolerance were essentially tacked on. In a way, this resembles what happened with Redis, where failover and clustering were introduced as an afterthought. However, unlike Redis, which chose to implement its own failover coordination and cluster-membership protocol, Solr opted to use ZooKeeper as an external coordinator.

We see that Elasticsearch’s core advantage is its distributed nature. Following that notion, it makes sense for it to own that coordination, which is why its designers chose to implement their own internal cluster membership, ZenDisco. But it turns out writing cluster-membership protocols is really fucking hard, and unless you’ve written proofs for it, you probably shouldn’t do it at all. The analogy here would be writing your own encryption algorithm—there’s tons of institutional knowledge which has laid the groundwork for solutions which are well-researched and well-understood. That knowledge should be embraced in situations like this.

I don’t mean to pick on Redis and Elasticsearch. They’re both excellent systems, but they serve as good examples for this discussion. The problem is that users of these systems tend to overlook the issues exposed by this mentality. Frankly, few people would know problems exist unless they are clearly documented by vendors (and not sales people) and even then, how many people actually read the docs cover-to-cover? It’s essential we know a system’s shortcomings and edge cases so we can recognize which situations to apply it and, more important, which we should not.

You don’t have to rely on an existing third-party library or service. Believe it or not, this isn’t a sales pitch for ZooKeeper. If it’s a core business function, it probably makes sense to build it yourself as Joel describes. What doesn’t make sense, however, is to build out whatever that is without being cognizant of conventional wisdom. I’m amazed at how often people are willing to throw away institutional knowledge, either because they don’t seek it out or they think they can do better (without formal verification). If I have seen further, it is by standing on the shoulders of giants.