Presentation on theme: "SDN Controller Challenges. The Story Thus Far SDN --- centralize the network’s control plane – The controller is effectively the brain of the network."— Presentation transcript:
SDN Controller Challenges
The Story Thus Far SDN --- centralize the network’s control plane – The controller is effectively the brain of the network – Controller determines what to do and tell switches how to do it.
The Story Thus Far
The Story Thus Far Let’s Ask the Brian!!!!
The Story Thus Far Think about what happen… Maybe come up with a solution Think about what happen… Maybe come up with a solution
The Story Thus Far Controller runs control function Control function creates switch state – F(global network state) Switch state – Global network state can be graph of the network Tell the network what to do
Challenges with Centralization Single point of failure – Fault tolerance Performance bottleneck – Scalability – Efficiency (switch-controller latency) Single point for security violations
Motivation for Distributed Controllers Wide-Area-Network – Wide distribution of switches: from USA to Australia. – High latency between one controller and All switches Application + Network growth – Higher CPU load for controller – More memory for storing FIB entries and calculations High availabilit
Class Outline Fault Tolerance – Google’s B4 paper Controller Scalability – Ways to scale the controller – Distributed controllers: Mesh Versus Hierarchy – Implications of controller placement
Google’s B4 Network Provides connectivity between DC sites Uses SDN to control edge switches Goal: high utilization of links Insight: fine-grained control over edge and network can lead to higher utilization Distributed Controllers – One set of controllers for each Data center (site)
Google’s B4 Network Provides connectivity between DC sites Uses SDN to control edge switches Goal: high utilization of links Distributed Controllers – One set of controllers for each Data center (site)
Fault Tolerance in B4 Each site runs a set of controller Paxos is run between controllers in a site to determine master
Quick Overview of Paxos Given N controllers – 1 Acts as leader, and N-1 as workers – All N controller maintain the same state Switches interact with leader Change doesn’t happen until whole group agrees Failure of primary N-1 work together to elect a new leader(determine new leader) Network Events Propagate State changes
Pros-Cons of Paxos Pros – Well understood and studied; gives good FT – Many implementations in the wild – E.g. Zookeeper Cons – Time to recover – Impacts through of the put of the entire system
What limits a controller’s scalability? Number of control messages from switch – Depends on the application logic E.g. MicroTE/Hedera periodically query all switches for stats Reactive controller, evaluated in NoX, requires each switch to send messages for a new flow – Packet-in (if reactive Apps) – Flow stats, Flow_time-outs
What limits a controller’s scalability? Application processing overhead The controller runs a bunch of application – Similar to: A server running a set of programs – CPU/Memory constraint limit how the app runs
What limits a controller’s scalability? Distance between controller and the switches Controller 1 Hedera L3 FW
How to Scale the Controller. Obvious: add more controllers. BUT: how about the applications? – Synchronization/concurrency problems. Who controls which switch? Who reacts to which events? Controller 1 Hedera L3 FW Controller 2 Hedera L3 FW Controller N Hedera L3 FW ? ? Stats + Install OF entries
Medium Sized Networks Assumption: – controller can’t store all forwarding table entries in memory – But can process all events and run all apps Each controller – Get same network events+ running same app. same output – But store output for only a fraction and config only a fraction Controller 1 Hedera L3 FW Controller 2 Hedera L3 FW Controller N Hedera L3 FW Stats + Install OF entries
Medium Sized Networks: hyperflow Each controller – Push state to each controller – Each controller things it’s the only one in the network Controller 1 Hedera L3 FW Controller 2 Hedera L3 FW Controller N Hedera L3 FW Stats + Install OF entries Sub-subscribe ssytem
Large Sized Networks Assumptions – Each controller can’t store all the FIB entries – Each controller can’t run the entire application or handle events Need to partition the application – But how?
Application partition 1 Approach 1: each controller runs a specific application – How do your resolve conflicts in FW entries – Apps can conflict in the rules they install Controller 1 Hedera Controller 2 L3 Controller N FW
Application partition 2 Approach 2: all controllers run the same application but for a subset of devices – Results in a Distributed Mesh control plane Controller 1 Hedera L3 FW Controller 2 Hedera L3 FW Controller N Hedera L3 FW Abstract Network view Abstract Network view
Application Partition 2 Abstract view exchanged with each other – Abstract view reduces the n/w information used by each controller Controller 2 Hedera L3 FW REAL NETWORK Controller 2’s View of NETWORK Abstraction Provided by Controller 1 Abstraction Provided by Controller N
ONIX to the SDN Programmer Controllers synchronize through a DB or DHT – So each app needs synchronization code. – How do you deal with concurrency. How to synchronize between domains. How many domains? Or controllers? How many switches in a domain?
Application partition 3 Approach 3: divide application into local, and global. – Results in a hierarchical control plane Global Controller and Local Controllers – Applications that do not need network-wide state Can be run locally without communicate with other controllers
Are Hierarchical Controllers Feasible Examples of local applications: – Link Discovery, Learning switch, local policies Examples of local portions of a global algo – Data center Traffic engineering Elephant flow detection (hedera) Predictability detection (MicroTE) Local apps/controllers have other benefits – High parallelism – Can be run closer to the devices.
Kandoo: Hierarchical controllers Controller 1 Hedera L3 FW Controller 2 Hedera L3 FW Controller N Hedera L3 FW Global Controller Hedera 2 levels of controllers: global and local – Local applications are embarrassingly parallel – Local shields global from network events
Kandoo: Hierarchical controllers Controller 1 Hedera L3 FW Controller 2 Hedera L3 FW Controller N Hedera L3 FW Global Controller Hedera Local Controllers: run local apps – Returns abstract view to the global controller – Reduces # events sent to global and reduce size of network seen by
Kandoo: Hierarchical controllers Controller 1 Hedera L3 FW Controller 2 Hedera L3 FW Controller N Hedera L3 FW Global Controller Hedera Global Controllers – Runs global apps: AKA apps that need network wide state
Hedera Reminder Goal: reduce network contention Insight: contention happens when elephants share paths. Solution: – Detect Elephant flows – Place Elephant flows on different flows
Implementing Hedera in Onix Controller 1 Hedera: detection +placement Hedera: detection +placement Controller 2 Hedera: detection+placement 2 levels of controllers: global and local – Local applications are embarrassingly parallel – Local shields global from network events Stats Flow Table entries Flow Table entries Flow Table entries Flow Table entries Exchange TM+detection
Implementing Hedera in Kandoo Controller 1 Elephant detection Controller 2 Controller N Global Controller Hedera: Global placement Local Controllers: get stats from networks + elephant detection Global Controller: decide flow placement + flow installation Elephant detection Inform of elephant flows Stats Install new flow table entries
Implementing B4 in Kandoo like architecture Site Controller Elephant detection Site Controller 2 Site Controller N Global Controller TE+BW allocator Local Controllers: get stats from networks + determines demand Global Controller: calculate paths for traffic Elephant detection Install TE Ops Stats + Install OF entries TE DB Inform of Flow demands
Kandoo to the SDN Programmer Think of what is local and what is global – When apps are written, annotate with local flag Kandoo will automatically place local – And place global. Kandoo restricts messages between global and local controllers – You can’t send OF styles messages – Must send Kandoo style messages
Summary Centralization provide simplicity at the cost of reliability and scalability Replication can improve reliability and scalability For Reliability, Paxos is an option For Scalability, conqueror and divide – Partition the applications Kandoo: Local apps and global apps – Partition the network Onix: each controller controls a subset of switches (Domain)