Day 5 Part 3: Elastic Distributed SDN Controller

Elastic distributed controller architecture can be described as the architecture in which the controller pool is dynamically grown or shrunk according to traffic conditions and the load is dynamically shifted across controllers.

This paper achieves the following:
  • A basic distributed controller architecture
  • A protocol that helps in switch migration for elasticity
The basic architecture proposed here is very similar to the one presented by the cluster based distributed controller paper. But here are the differences:
  • Each switch has many equal controllers and many slave controllers
  • Each controller node maintains a TCP channel with every other controller
  • A controller may use this channel to send a message to a switch connected to another controller
  • This is how the migration is made easier
  • application state is maintained in each controller to facilitate switch migration and fail over recovery

The Unique Switch Migration Mechanism

The mechanism is as follows:
  • check and make the new master controller as an equal controller if it wasn't already
  • current master controller adds a dummy flow and deletes it. When the flow gets deleted, the delete message goes to both current and new master controller and thus communication as equals is established
  • something called a barrier request and reply is made between the current master and switch to say that they will end communication as master - switch. 
  • a role request from the new master is sent to switch to establish the fact that it is the new master
The normal way of first associating the new master as equal and then terminating the old master controller communication might not work. This is because once the new master takes over, the switch can override the commands sent by the old master controller.

Comments

  1. Ram P RustagiFebruary 2, 2019 at 1:02 AM

    As I understood from earlier description, the job of master controller is just to keep mapping of switches to controllers. If it is keeping flow table entries, then I am unclear on how the hierarchy will work as this master controller would become a bottleneck in terms of resources/

    ReplyDelete
    Replies
    1. ShravanyaFebruary 2, 2019 at 7:45 AM

      Earlier I had thought that master would maintain only the mapping. But since communication from one controller domain to another is not possible without another controller that knows the whole network, I have added the feature to the master controller from here on. So, the conclusion is: yes, the master controller will receive OpenFLow PCK_IN in cases of cross controller domain. This is the reason it will act like any other controller in the architecture. Only, it has access to all switches and not restricted to one domain.

      Delete
    2. ShravanyaFebruary 2, 2019 at 7:45 AM

      In the further articles, I have accounted for bottlenecks in the master controller as well.

      Delete

Post a Comment

Popular posts from this blog

Day 12: Master Slave SDN Controller Architecture

In the previous post, I was struggling to figure out why all controllers can install flows in all switches irrespective of the code. Well, today the mystery is getting resolved. I found this book that explains beautifully the various roles that a controller can play. The three roles are: Equal : this happens by default all the equal controllers in a network that are connected to the switch can add and delete flows from it Slave: it has a read-only limitation to the switch openflow tables Master: very similar to the switch, but only 1 master can be connected to the switch at that point of time, and no other flows can intercept the communication between switch and master. I highly recommend you go through the above link to get a better picture. Since the default role of any controller is "Equal", we couldn't really differentiate between the three controller flows yesterday. But one good news from this is that we don't really need extensive coding to assi...
Read more

Day 1: Understanding Ransomware and how to detect them?

Image
To start this journey, let’s begin at the basics and learn about what ransomwares are, what is the meaning of a ransomware attack, how have other people detected it so far? By learning the above, we will be in a position to analyze any ransomware attack and devise methods to detect, mitigate and prevent them. Definition: Ransomware is a type of malicious software from cryptovirology that threatens to publish the victim's data or perpetually block access to it unless a ransom is paid. To continue understanding and addressing the threats posed by ransomware, let us dig a little deeper. As the definition specifies, ransomware is a type of malicious software. Now, what is a malicious software? Definition: Malware (= Malicious + Software) is a software which is specifically designed to disrupt, damage, or gain authorized access to a computer system. These malwares (or simply dangerous content that can corrupt your system), can enter your system from various s...
Read more

Day 50: Tcpreplay and tcpliveplay approach

Today we shall be exploring how to copy and redirect the OpenFlow packets to more than one controller from the load balancer. The load balancer is anyway sending the OpenFlow packets to one of the slave controllers - c1 or c2. The packets sent to these two controllers needs to be sent to master controller so the master controller can update the new flows added by c1 or c2 on the switches. This ensures synchronization between controllers. There are two implementation methods to achieve the same: Copy all packets generated from the load balancer to the controller side. Only the packets having destination IP address of c1 and c2 can be redirected to the master controller. Thus we can prevent the packets which already are reaching the master controller to be sent as duplicates Copy and redirect packets from the slave controllers - c1 and c2. We need not take care of duplicate packets reaching master controller. This will reduce the time taken to redirect packets. Since the second a...
Read more