Riak Core: Building Distributed Applications Without Shared State

Riak Core: Building Distributed Applications Without Shared State Commercial Users of Functional Programming Baltimore, MD · October 2010 Rusty Klophaus (@rklophaus) Basho Technologies Wednesday, October 6, 2010

You suddenly feel an uncontrollable desire to learn Erlang. http://www.ﬂickr.com/photos/procsilas/18014203 2 Wednesday, October 6, 2010

What is Riak Core? How does it work? How can you use it? 3 Wednesday, October 6, 2010

Distributed, scalable, failure-tolerant. 4 Wednesday, October 6, 2010

Distributed, scalable, failure-tolerant. No central coordinator. Easy to setup/operate. 5 Wednesday, October 6, 2010

Distributed, scalable, failure-tolerant. Horizontally scalable; add commodity hardware to get more X. 6 Wednesday, October 6, 2010

Distributed, scalable, failure-tolerant. Always available. No single point of failure. Self-healing. 7 Wednesday, October 6, 2010

Basho Technologies Riak KV Distributed, scalable, failure-tolerant key/value datastore. Started as a “Dynamo clone”. Map/Reduce, Lightweight Data Relations, Client APIs Riak Search Distributed, scalable, failure-tolerant full-text search engine. Near Realtime, Riak KV Integration, Solr Support 8 Wednesday, October 6, 2010

Wednesday, October 6, 2010

Riak Riak Riak KV Core Search Wednesday, October 6, 2010

Riak Core is an Erlang library that helps you build distributed, scalable, failure-tolerant applications. 11 Wednesday, October 6, 2010

Amazon Dynamo Wednesday, October 6, 2010

“We Generalized the Dynamo Architecture and Open-Sourced the Bits.” 13 Wednesday, October 6, 2010

Wait, doesn’t *Erlang* let you build distributed, scalable, failure-tolerant applications? 14 Wednesday, October 6, 2010

Erlang makes it easy to connect the components of your application. Client Service A Service B Service C Queue E Resource D Wednesday, October 6, 2010

Riak Core helps you build a service that harnesses the power of many nodes. Node A Node B Node C Node D Node E Node F Node G Node H Service Node I Node J Node K Node L Node M Node N Node O ... Wednesday, October 6, 2010

“People use languages other than Erlang?!? I ﬁnd that hilarious.” -Al Gore actually said this to me. http://www.ﬂickr.com/photos/scobleizer/2216445692 Wednesday, October 6, 2010

How does Riak Core work? 18 Wednesday, October 6, 2010

Command ObjectName, Payload Wednesday, October 6, 2010

Predictable Routing 20 Wednesday, October 6, 2010

Hash the Object Name Command ObjectName, Payload SHA1(ObjName), Payload 0 to 2^160 Wednesday, October 6, 2010

A Naive Approach Command ObjectName, Payload SHA1(ObjName), Payload Node A Node B Node C Node D Wednesday, October 6, 2010

A Naive Approach Command ObjectName, Payload SHA1(ObjName), Payload Node A Node B Node C Node D Node E Wednesday, October 6, 2010

"All problems in computer science can be solved by another level of indirection." - David Wheeler 24 Wednesday, October 6, 2010

Add VNodes What Virtual Node. Logical subdivision of the cluster. Handles incoming commands, does work, replies. For Parallelism # of VNodes = maximum concurrent requests For Rebalancing the Cluster Smallest block that can be shifted to a new node. For Resiliance The system restarts failed VNodes. 25 Wednesday, October 6, 2010

Routing with Consistent Hash Command ObjectName, Payload SHA1(ObjName), Payload VNode 0 VNode 1 VNode 2 VNode 3 VNode 4 VNode 5 VNode 6 VNode 7 Node A Node B Node C Node D Wednesday, October 6, 2010

Adding a Node Command ObjectName, Payload SHA1(ObjName), Payload VNode 0 VNode 1 VNode 2 VNode 3 VNode 4 VNode 5 VNode 6 VNode 7 Node A Node B Node C Node D Node E Wednesday, October 6, 2010

Removing a Node Command ObjectName, Payload SHA1(ObjName), Payload VNode 0 VNode 1 VNode 2 VNode 3 VNode 4 VNode 5 VNode 6 VNode 7 Node A Node B Node C Node D Node E Wednesday, October 6, 2010

The Ring Hash Location Wednesday, October 6, 2010

The Ring Preﬂist Wednesday, October 6, 2010

Writing Replicas (n_val) Preﬂist when N=3 Wednesday, October 6, 2010

Routing Around Failures X Preﬂist when N=3 and node 0 is down. Wednesday, October 6, 2010

Location of the Routing Layer 33 Wednesday, October 6, 2010

Router in the Middle Client Client Client Router VNode VNode VNode VNode VNode VNode VNode VNode 0 1 3 4 2 5 6 7 Node A Node B Node C Node D Node E Wednesday, October 6, 2010

Riak Core - Router on Each Node Client Client Client Router Router Router Router Router VNode VNode VNode VNode VNode VNode VNode VNode 0 1 3 4 2 5 6 7 Node A Node B Node C Node D Node E Wednesday, October 6, 2010

Eventually - Router on the Client Client Client Client Router Router Router VNode VNode VNode VNode VNode VNode VNode VNode 0 1 3 4 2 5 6 7 Node A Node B Node C Node D Node E Wednesday, October 6, 2010

No Shared State Router Router Router Router Router VNode VNode VNode VNode VNode VNode VNode VNode 0 1 3 4 2 5 6 7 Node A Node B Node C Node D Node E Wednesday, October 6, 2010

Gossip Local Incoming Ring State Ring State Are rings equivalent? Strictly descendent? Or different? Wednesday, October 6, 2010

Handoff When Nodes is added to the system. Node is removed from the system. Node has temporarily failed. What Ship the data backing a VNode from one node to another. 39 Wednesday, October 6, 2010

Not Mentioned Vector Clocks Merkle Trees Bloom Filters 40 Wednesday, October 6, 2010

Distinguished gentlemen prefer Erlang. http://www.ﬂickr.com/photos/rebcal/3987226359 Wednesday, October 6, 2010

How do you use Riak Core? 42 Wednesday, October 6, 2010

Two Things to Think About Command Command = ObjectName, Payload The commands/requests/operations that you will send through the system. VNode Module The callback module that will receive the commands. 43 Wednesday, October 6, 2010

VNode Module Startup/Shutdown init([Partition]) -> {ok, State} terminate(State) -> ok Commands handle_command(Cmd, Sender, State) -> {noreply, State1} | {reply, Reply, State1} handle_handoff_command(Cmd, Sender, State) -> {noreply, State1} | {reply, ok, State1} 44 Wednesday, October 6, 2010

VNode Module Handoff Coordination handoff_starting(Node, State) -> {Bool, State1} encode_handoff_data(Data, State) -> <<Binary>>. handle_handoff_data(Data, Sender, State) -> {reply, ok, State1} handoff_finished(Node, State) -> {ok, State1} 45 Wednesday, October 6, 2010

Start the riak_core application riak_core riak_core_vnode_sup riak_core_handoff_* riak_core_node_* X_vnode X_vnode riak_core_ring_* riak_core_gossip_* X_vnode ... application:start(riak_core). 46 Wednesday, October 6, 2010

Start the riak_core application riak_core riak_core_vnode_sup riak_core_handoff_* riak_core_node_* X_vnode X_vnode riak_core_ring_* riak_core_gossip_* X_vnode ... Supervise vnode processes. 47 Wednesday, October 6, 2010

Start the riak_core application Start, coordinate, and supervise handoff. 48 Wednesday, October 6, 2010

Start the riak_core application riak_core riak_core_vnode_sup riak_core_handoff_* riak_core_node_* X_vnode X_vnode riak_core_ring_* riak_core_gossip_* X_vnode ... Maintain cluster membership information. 49 Wednesday, October 6, 2010

Start the riak_core application riak_core riak_core_vnode_sup riak_core_handoff_* riak_core_node_* X_vnode X_vnode riak_core_ring_* riak_core_gossip_* X_vnode ... Monitor node liveness, broadcast to registered modules. 50 Wednesday, October 6, 2010

Start the riak_core application riak_core riak_core_vnode_sup riak_core_handoff_* riak_core_node_* X_vnode X_vnode riak_core_ring_* riak_core_gossip_* X_vnode ... Send ring information to other nodes. Reconcile different views of the cluster. Rebalance cluster when nodes join or leave. 51 Wednesday, October 6, 2010

In your application... riak_core riak_core_vnode_sup riak_core_handoff_* riak_core_node_* X_vnode X_vnode riak_core_ring_* riak_core_gossip_* X_vnode ... Start the vnodes for your application. Master = { riak_X_vnode_master, { riak_core_vnode_master, start_link, [riak_X_vnode] }, permanent, 5000, worker, [riak_core_vnode_master] }, {ok, { {one_for_one, 5, 10}, [Master]} }. 52 Wednesday, October 6, 2010

In your application... riak_core riak_core_vnode_sup riak_core_handoff_* riak_core_node_* X_vnode X_vnode riak_core_ring_* riak_core_gossip_* X_vnode ... Tell riak_core that your application is ready to receive requests. riak_core:register_vnode_module(riak_X_vnode), riak_core_node_watcher:service_up(riak_X, self()) 53 Wednesday, October 6, 2010

In your application... riak_core riak_core_vnode_sup riak_core_handoff_* riak_core_node_* X_vnode X_vnode riak_core_ring_* riak_core_gossip_* X_vnode ... riak_core riak_core_vnode_sup riak_core_handoff_* riak_core_node_* X_vnode X_vnode riak_core_ring_* riak_core_gossip_* X_vnode ... Join to an existing node in the cluster. riak_core_gossip:send_ring(ClusterNode, node()) 54 Wednesday, October 6, 2010

Start Sending Commands # Figure out the preflist... {_Verb, ObjName, _Payload} = Command, PrefList = riak_core_apl:get_apl(ObjName, NVal, riak_X), # Send the command... riak_core_vnode_master:command(PrefList, Command, riak_X_vnode_master) 55 Wednesday, October 6, 2010

Review Riak Core Open source Erlang library for building distributed, scalable, failure tolerant applications. Continual improvement in the coming months. Riak KV Key/Value datastore with map/reduce based on Riak Core. Riak Search Full-text, near real-time search engine based on Riak Core. 56 Wednesday, October 6, 2010

At the center of the universe, beneath the bottom-most turtle, is a cluster of Erlang nodes. http://www.ﬂickr.com/photos/jurvetson/469492885 Wednesday, October 6, 2010

At the center of the universe, beneath the bottom-most turtle, is a cluster of Erlang nodes. With 100% uptime. http://www.ﬂickr.com/photos/jurvetson/469492885 Wednesday, October 6, 2010

Thanks! Questions? Learn More More Information: http://wiki.basho.com Amazon’s Dynamo Paper Get the Code http://hg.basho.com/riak_core Get in Touch Twitter: @rklophaus, @basho/team Email: rusty@basho.com 59 Wednesday, October 6, 2010

END Wednesday, October 6, 2010

http://nosql.mypopescu.com	322
http://www.nosqldatabases.com	58
http://irr.posterous.com	16
http://static.slidesharecdn.com	4
http://www.linkedin.com	2
resource://brief-content	1
applewebdata://62A4CE7A-4DD3-4693-A9A0-E094EF82A752	1
http://www.docshut.com	1

Riak Core: Building Distributed Applications Without Shared State

by Rusty Klophaus on Oct 12, 2010

Accessibility

Categories

Upload Details

Usage Rights

8 Embeds 405

Statistics

Riak Core: Building Distributed Applications Without Shared State Presentation Transcript