Are folks OK with the current iteration of this being in the 0.7.1 beta? It would be in a separate module, so no additional dependencies are required, and it could be a good opportunity to get some more feedback.Date: Fri, 22 Aug 2014 13:42:31 -0700 Subject: Re: Add performant IPC (Helix actors) From: g.kish...@gmail.com To: dev@helix.apache.org CC: u...@helix.apache.org
Thanks Greg for explanation. I agree Akka is 2 or 3 layers above what Helix-ipc is trying to provide. Also there are some nuances in order guarantees with Akka. For control messages where reliability is a must and probably more important than latency, I think using ZK as the communication channel is the right solution. Without pushing state transition message through a consensus system like ZK, there will be lot of edge cases where ensuring correctness becomes impossible. On Fri, Aug 22, 2014 at 1:29 PM, Greg Brandt <brandt.g...@gmail.com> wrote: Hey Henry, So Akka actors actually give exactly the same messaging guarantees as the current implementation of Helix IPC: at-most-once message delivery, and ordering per sender-receiver pair. In Helix's case, the latter can be thought of further as ordering per-partition or resource, but basically is guaranteed by Netty's Channel implementation for sender-receiver pair ( http://stackoverflow.com/questions/9082101/do-messages-sent-via-nettys-channel-write-preserve-their-order-when-begin-sen ). What Akka provides that this IPC layer doesn't is the application life cycle management part. This is why I made the claim that Akka Actors are analogous to Helix Participants. One writes callbacks for different message types in the Actor API to manage various control messages in the same way one writes callbacks in Helix to manage application state transitions. But more importantly, using Akka instead of getting really down into the transport layer might hide too much when we want to use it for performance-critical things like data replication. Continuing with that example, we likely want to use direct IO and minimize data copies in the transport layer, and it seems like some of the niceness involved in Akka's API would get in the way of that. Something that may be really cool to do is write another Akka transport implementation on top of this, so you could basically add partition/state to the Akka API and not have to implement the state machines yourself (as is shown in some Akka docs). In that scenario, users should probably be unaware that they were using Helix and just use the same Akka API. -Greg On Fri, Aug 22, 2014 at 11:13 AM, Henry Saputra <henry.sapu...@gmail.com> wrote: > Hi Greg, > > Thanks for the reply. Sorry for late for the discussions. > > If the IPC semantic will be used only by Helix internally I agree that > lower level transport like Netty should be the way to go. > > For control plane layer, such as communications between controller and > participants, wouldn't Akka actors give benefits of already reliable > message passing and monitoring without building from ground up? > Akka actor system has many different routers and allow callback to > caller to get async response (in this case return back response to > controller). > > I just saw the similarity of what the new API looks like with what > Akka actor APIs. > > Thanks, > > - Henry > > On Wed, Aug 20, 2014 at 5:46 PM, Greg Brandt <brandt.g...@gmail.com> > wrote: > > Hey Henry, > > > > Initially, I thought the same thing. But after evaluating Akka Actors, I > > think they're a fundamentally different approach to developing > distributed > > systems as opposed to simply a messaging layer implementation. Let me > > explain... > > > > The primary difference between Akka Actors and Helix is that the former > > prefers loose, decentralized control, whereas the latter prefers strong, > > centralized control. Akka applications are expected to manage state by > > themselves, whereas in Helix, that responsibility is delegated to the > > controller. And the actual user Akka Actor implementations are analogous > to > > Helix Participant implementations. > > > > So with that perspective, it would be a bit of a kludge to leverage Akka > > Actors for simple cluster-scope-addressable message passing. Actually, > > Helix already provides this messaging feature, but it is done via > ZooKeeper > > which is dangerous and less-than-performant for a high volume of > messages. > > > > The right thing to do seems to be to provide a set of better transport > > implemenations (e.g. Netty, ActiveMQ, Kafka, etc.) for the > > cluster-scope-addressable messaging system, and better preserve the > > fundamental concepts in Helix. > > > > Let me know what you think. > > > > -Greg > > > > > > > > On Wed, Aug 20, 2014 at 9:55 AM, Henry Saputra <henry.sapu...@gmail.com> > > wrote: > > > >> This seems fitting for Akka actor system [1]. Maybe we could just use > >> it as base? > >> > >> > >> - Henry > >> > >> [1] http://akka.io > >> > >> On Mon, Jul 21, 2014 at 9:47 AM, Greg Brandt <brandt.g...@gmail.com> > >> wrote: > >> > We talked about this in a little bit greater detail, and it's seeming > >> like > >> > there are three basic components that we want: > >> > > >> > 1. Transport - physically moving messages > >> > 2. Resolution - map cluster scope to set of machines > >> > 3. Management - message state, errors, retries, etc. > >> > > >> > The transport interface should be generic enough to allow many > >> > implementations (e.g. Netty, Kafka, etc.) It may be valuable to do > >> multiple > >> > implementations if one underlying system supports desired message > >> delivery > >> > guarantees better than another. For example, Kafka would be more > amenable > >> > to at-least-once delivery, whereas something like Netty, with much > less > >> > overhead than Kafka, would be better for at-most-once delivery. > >> > > >> > The goal here I think is to provide something that's most similar to > Akka > >> > Actors, with the following guarantees: > >> > > >> > 1. at-most-once delivery > >> > 2. message ordering per sender-receiver pair > >> > > >> > A light-weight messaging primitive with those guarantees would allow > >> > implementation of many interesting things like Kishore mentions. Netty > >> > seems most suited to this task (it is the underlying transport for > Akka > >> as > >> > well). The reason to avoid Akka I think is that it imposes too much of > >> its > >> > own philosophy on the user, and is really bloated. A Netty-based > >> primitive > >> > could be provided out-of-the-box, and the interfaces it implements > should > >> > be easy enough for users to implement on their own. It would be > >> relatively > >> > trivial to add a Kafka-based implementation out-of-the-box as well. > >> > > >> > (Have a bit of a prototype here: > >> > > >> > https://github.com/brandtg/helix-actors/blob/master/helix-actors/src/main/java/org/apache/helix/actor/netty/NettyHelixActor.java > >> > ) > >> > > >> > The transport mechanism should be agnostic to anything going on in the > >> > cluster: the resolution module is totally separate, and maps cluster > >> scope > >> > to sets of nodes. For example, one could send a message to the entire > >> > cluster, specific partitions of a resource, or even from one cluster > to > >> > another (via composing resolution modules or something). > >> > > >> > Then there would be a management component, composed with the other > two > >> > components, which manages user callbacks associated with varying > cluster > >> > scopes, tracks messages based on ID in order to perform request / > >> response > >> > or feedback-loop kinds of things, manages errors / timeouts / retries, > >> etc. > >> > > >> > -Greg > >> > > >> > > >> > > >> > On Sat, Jul 12, 2014 at 11:27 AM, kishore g <g.kish...@gmail.com> > wrote: > >> > > >> >> Hi Vlad, > >> >> > >> >> Yes the idea is to have a pluggable architecture where Helix provides > >> the > >> >> messaging channel without serializing/deserializing the data. In the > >> end, I > >> >> envision having most of the building blocks needed to build a > >> distributed > >> >> system. One should be able to build a datastore/search/olap/pu > >> sub/stream > >> >> processing systems easily on top of Helix. > >> >> > >> >> I am curious to see what kind of tuning you guys did to handle 1.5 > QPS. > >> Our > >> >> goal is to have 1M QPS between the two servers. In order to achieve > >> this we > >> >> might need pipelining/batching along with ordering guarantee's. This > is > >> non > >> >> trivial and error prone, I see a huge value in providing this > feature. > >> Most > >> >> systems have built some form replication schemes on their own. There > are > >> >> two parts in the replication, one is the high through put message > >> exchange > >> >> and the other is the scheme that is synchronous replication/async > >> >> replication/chained replication/consensus etc. The first part is the > >> common > >> >> part that is needed across all the replication scheme. > >> >> > >> >> Of course getting this right is not easy and design is tricky but I > feel > >> >> its worth a try. > >> >> > >> >> This is the high level flow I had discussed with Greg. > >> >> > >> >> Each server can host P partitions and is listening at one port. > Based on > >> >> how many execution threads are needed, we can create C channels. > Every > >> >> partition maps to a specific channel. On each channel we can > guarantee > >> >> message ordering. We simply provide a callback to handle each message > >> and > >> >> pass the received bytes without deserializing. Its up to the > callback to > >> >> either process it synchronously or asynchronously. > >> >> > >> >> There are more details that its difficult to go over in the email. We > >> >> should probably write up the design and then deep dive on the > details. > >> >> > >> >> thanks, > >> >> Kishore G > >> >> > >> >> > >> >> > >> >> > >> >> > >> >> > >> >> > >> >> > >> >> > >> >> > >> >> > >> >> > >> >> > >> >> > >> >> > >> >> > >> >> On Sat, Jul 12, 2014 at 8:08 AM, vlad...@gmail.com < > vlad...@gmail.com> > >> >> wrote: > >> >> > >> >> > I agree that having a messaging scheme that is not > ZooKeeper-reliant > >> >> would > >> >> > help in building some synchronization protocols between clients and > >> >> > replicas. I am thinking of cases such as writing inside Kafka to > >> >> replicated > >> >> > partitions our our own application,a replicated Key/Value store. > >> >> > > >> >> > For me, service discovery bridges the gap between an application > >> wanting > >> >> > to address a service (in Helix's case more particularly, a > >> >> service/resource > >> >> > name/partition tuple) and finding out a list of hostnames that can > >> >> receive > >> >> > that request. After that, whether the application wants to use > Helix > >> >> itself > >> >> > or another messaging stack is an open problem, perhaps a clear > >> separation > >> >> > of the discovery/messaging layers would allow for quickly swapping > >> >> > different pieces here. > >> >> > > >> >> > The interesting part (in my view) in providing fast messaging as > part > >> of > >> >> > Helix is the ability to use the service topology information that > >> Helix > >> >> > hosts (number of replicas and their placement) together with the > >> >> messaging > >> >> > layer as sufficient components for realizing a consensus algorithm > >> >> between > >> >> > replicas, that is providing, on top of the regular service > discovery > >> >> > function of writing to one instance of a service, the function of > >> writing > >> >> > with consensus verification to all replicas of a service. > >> >> > > >> >> > In terms of capabilities, looking at Kafka and KV/Store as use > cases, > >> >> this > >> >> > would mean running a whole protocol between the replicas of the > >> service > >> >> to > >> >> > achieve a certain synchronization outcome (be it Paxos, 2 Phase > >> Commit or > >> >> > Chain Replication - Paxos would probably not be needed since Helix > >> >> already > >> >> > provides a resilient master for the partition, so there is a clear > >> >> > hierarchy). Kafka suggests even more cases if we take the > interaction > >> >> > between client and service into account, namely specifying the > call to > >> >> > return after the master gets the data or after consensus between > >> master > >> >> and > >> >> > replicas have been achieved, or sending data asynchronously (the > >> >> > combination of the producer.type and request.required.acks > parameters > >> of > >> >> > the producer). > >> >> > > >> >> > The only reason for which I would like to see a pluggable > >> architecture is > >> >> > that in high volume applications the messaging stack requires a > lot of > >> >> > tuning. As I said at the meet-up, in our company we deal with about > >> 1.5M > >> >> > QPS, so having serialization/deserialization overheads per > message, or > >> >> > event separate listening/executor threads can be resource > intensive. > >> >> > Whenever we can, we are tempted to use a stage-event drive > >> architecture, > >> >> > with few threads that handle large message pools. This also raises > the > >> >> > question of how to implement the upper-layer callbacks (that would > >> >> perform > >> >> > the synchronization-related actions) without spawning new threads. > >> >> > > >> >> > Regards, > >> >> > Vlad > >> >> > > >> >> > > >> >> > On Fri, Jul 11, 2014 at 11:40 PM, kishore g <g.kish...@gmail.com> > >> wrote: > >> >> > > >> >> >> Good point Vlad, I was thinking of defining the right abstraction > and > >> >> >> possibly provide one implementation based. I think the synchronous > >> and > >> >> >> asynchronous messaging should be covered as part of this. > >> >> >> > >> >> >> Also I think Finagle and likes of it cater more towards the client > >> >> server > >> >> >> communication but what we lack today is a good solution for peer > to > >> peer > >> >> >> transport. For example, if some one has to build a consensus layer > >> they > >> >> >> have build everything from ground up. Providing a base layer that > >> >> exchange > >> >> >> messages between peers on a per partition basis can be a great > >> building > >> >> >> block to build different replication schemes. > >> >> >> > >> >> >> Overall messaging can be used for two cases > >> >> >> -- data and control. > >> >> >> > >> >> >> > >> >> >> CONTROL > >> >> >> > >> >> >> This is needed for control messages that occur rarely, for these > >> type of > >> >> >> messages latency/throughput is not important but its important > for it > >> >> to be > >> >> >> reliable. > >> >> >> > >> >> >> > >> >> >> DATA > >> >> >> > >> >> >> This is mainly exchanging data between different roles, > >> >> >> controller-participant, participant-participant, > >> spectator-participant. > >> >> >> These types of message exchanges occur quite frequently and having > >> high > >> >> >> throughput and low latency is a requirement. > >> >> >> > >> >> >> For example having the following api and guarantee would allow > one to > >> >> >> build synchronous replication, asynchronous, quorum etc. > >> >> >> > >> >> >> send(partition, message, ACK_MODE, callback) ACK_MODE can be like > ACK > >> >> >> from ALL, QUORUM, 1, NONE etc. callback is fired when one gets the > >> >> response > >> >> >> back from the receiver. > >> >> >> > >> >> >> This simple api can allow both synchronous and asynchronous mode > of > >> >> >> communication. We dont have to do the > serialization/deserialization. > >> The > >> >> >> hard part here would be to guarantee ordering between the > messages. > >> One > >> >> >> should be able to specify the message ordering requirement, FIFO > on a > >> >> per > >> >> >> partition level or dont care about ordering. Having this makes it > >> easy > >> >> for > >> >> >> one to implement replication schemes. > >> >> >> > >> >> >> thanks, > >> >> >> Kishore G > >> >> >> > >> >> >> > >> >> >> On Fri, Jul 11, 2014 at 12:49 PM, Greg Brandt < > brandt.g...@gmail.com > >> > > >> >> >> wrote: > >> >> >> > >> >> >>> Vlad, > >> >> >>> > >> >> >>> I'm not sure that the goal here is to cover all possible use > cases. > >> >> This > >> >> >>> is > >> >> >>> intended as basically a replacement for the current Helix cluster > >> >> >>> messaging > >> >> >>> service, which doesn't rely on ZooKeeper. I would argue that > Helix > >> is > >> >> >>> already that general framework for discovering the service > endpoint, > >> >> and > >> >> >>> that this is just one implementation of an underlying messaging > >> stack > >> >> >>> (for > >> >> >>> which there is currently only the ZooKeeper-based > implementation). > >> >> >>> > >> >> >>> Moreover, keeping it too general (i.e. providing the APIs > outlined > >> in > >> >> the > >> >> >>> JIRA, but no implementation) puts much greater onus on the user. > It > >> >> would > >> >> >>> be nice for someone just starting out with Helix to have a pretty > >> good > >> >> >>> messaging service readily available, as well as the APIs to > >> implement > >> >> >>> more > >> >> >>> specific solutions if he or she so chooses. > >> >> >>> > >> >> >>> -Greg > >> >> >>> > >> >> >>> > >> >> >>> On Fri, Jul 11, 2014 at 11:07 AM, vlad...@gmail.com < > >> vlad...@gmail.com > >> >> > > >> >> >>> wrote: > >> >> >>> > >> >> >>> > This sounds like service discovery for a messaging solution. At > >> this > >> >> >>> point > >> >> >>> > there would be some overlap with message buses such as Finagle. > >> >> Perhaps > >> >> >>> > this should be a general framework for discovering the service > >> >> endpoint > >> >> >>> > that could leave the actual underlaying messaging stack open > (be > >> it > >> >> >>> > Finagle, Netty-based or ZeroMQ, for example). My only fear is > >> that if > >> >> >>> the > >> >> >>> > messaging framework is encapsulated completely into Helix, it > >> would > >> >> be > >> >> >>> hard > >> >> >>> > to compete with tuned messaging bus solutions and cover all > >> possible > >> >> >>> use > >> >> >>> > cases (for example, in my company, we use a large number of sub > >> cases > >> >> >>> on > >> >> >>> > the synchronous call to asynchronous call spectrum). > >> >> >>> > > >> >> >>> > Regards, > >> >> >>> > Vlad > >> >> >>> > > >> >> >>> > > >> >> >>> > On Fri, Jul 11, 2014 at 10:56 AM, Greg Brandt < > >> brandt.g...@gmail.com > >> >> > > >> >> >>> > wrote: > >> >> >>> > > >> >> >>> >> (copied from HELIX-470) > >> >> >>> >> > >> >> >>> >> Helix is missing a high-performance way to exchange messages > >> among > >> >> >>> >> resource partitions, with a user-friendly API. > >> >> >>> >> > >> >> >>> >> Currently, the Helix messaging service relies on creating many > >> nodes > >> >> >>> in > >> >> >>> >> ZooKeeper, which can lead to ZooKeeper outages if messages are > >> sent > >> >> >>> too > >> >> >>> >> frequently. > >> >> >>> >> > >> >> >>> >> In order to avoid this, high-performance NIO-based HelixActors > >> >> should > >> >> >>> be > >> >> >>> >> implemented (in rough accordance with the actor model). > >> HelixActors > >> >> >>> exchange > >> >> >>> >> messages asynchronously without waiting for a response, and > are > >> >> >>> >> partition/state-addressable. > >> >> >>> >> > >> >> >>> >> The API would look something like this: > >> >> >>> >> > >> >> >>> >> public interface HelixActor<T> { > >> >> >>> >> void send(Partition partition, String state, T message); > >> >> >>> >> void register(String resource, HelixActorCallback<T> > >> callback); > >> >> >>> >> } > >> >> >>> >> public interface HelixActorCallback<T> { > >> >> >>> >> void onMessage(Partition partition, State state, T > message); > >> >> >>> >> } > >> >> >>> >> > >> >> >>> >> #send should likely support wildcards for partition number and > >> >> state, > >> >> >>> or > >> >> >>> >> its method signature might need to be massaged a little bit > for > >> more > >> >> >>> >> flexibility. But that's the basic idea. > >> >> >>> >> > >> >> >>> >> Nothing is inferred about the format of the messages - the > only > >> >> >>> metadata > >> >> >>> >> we need to be able to interpret is (1) partition name and (2) > >> state. > >> >> >>> The > >> >> >>> >> user provides a codec to encode / decode messages, so it's > nicer > >> to > >> >> >>> >> implementHelixActor#send and HelixActorCallback#onMessage. > >> >> >>> >> > >> >> >>> >> public interface HelixActorMessageCodec<T> { > >> >> >>> >> byte[] encode(T message); > >> >> >>> >> T decode(byte[] message); > >> >> >>> >> } > >> >> >>> >> > >> >> >>> >> Actors should support somewhere around 100k to 1M messages per > >> >> second. > >> >> >>> >> The Netty framework is a potential implementation candidate, > but > >> >> >>> should be > >> >> >>> >> thoroughly evaluated w.r.t. performance. > >> >> >>> >> > >> >> >>> > > >> >> >>> > > >> >> >>> > >> >> >> > >> >> >> > >> >> > > >> >> > >> >
