S. Veeramani

Bio: S. Veeramani is an academic researcher from Indian Institutes of Information Technology. The author has contributed to research in topics: Forwarding plane & Routing table. The author has an hindex of 1, co-authored 5 publications receiving 15 citations.

Minimization of flow table for TCAM based openflow switches by virtual compression approach

Abstract: Forwarding lookup in the open flow switch can be done for each arriving packet by every switch in path. Every switch maintains a set of IP prefix value in its lookup table. For a given IP address it finds a longest prefix match in the forwarding table that matches first fewer bits of destination address. Earlier, there are several technique that has been proposed to optimize the table space and to reduce the number of entries in the forwarding table. In TCAM prefix values needs to be stored in descending order of its length, it require more number of entries needs to be shifted. In this paper we propose a method to reduce the length of the prefix value up to 50 to 60% horizontally as well as vertically which will be stored in the forwarding table. It involves significant prefix methods to reduce number of entries in the forwarding table.

Reinforcement learning based path planning of multiple agents of SwarmItFIX robot for fixturing operation in sheetmetal milling process

Abstract: SwarmItFIX (self-reconfigurable intelligent swarm fixtures) is a multi-agent setup mainly used as a robotic fixture for large Sheet metal machining operations. A Constraint Satisfaction Problem (CSP) based planning model is utilized currently for computing the locomotion sequence of multiple agents of the SwarmItFIX. But the SwarmItFIX faces several challenges with the current planner as it fails on several occasions. Moreover, the current planner computes only the goal positions of the base agent, not the path. To overcome these issues, a novel hierarchical planner is proposed, which employs Monte Carlo and SARSA TD based model-free Reinforcement Learning (RL) algorithms for the computation of locomotion sequences of head and base agents, respectively. These methods hold two distinct features when compared with the existing methods (i) the transition model is not required for obtaining the locomotion sequence of the computational agent, and (ii) the state-space of the computational agent become scalable. The obtained results show that the proposed planner is capable of delivering optimal makespan for effective fixturing during the sheet metal milling process.

Hybrid trie based partitioning of TCAM based openflow switches

Abstract: The packet forwarding mechanism in the open flow network switch is based on the IP lookup in the forwarding table, which consist of the destination IP address of the incoming packet. In general, these forwarding tables are Content Addressable Memory (CAM), where the desire key will be searched in the table to know out going port. Ternary content addressable memory (TCAM) is one of the efficient mechanism to store the forwarding table, where values are stored in sorted order. Search of any key is done by using longest prefix matching technique. It uses 0, 1 and x (don't care) to represent the data, rather using 0 and 1 in case of CAM. The major drawback of TCAM is that it is a power hungry circuit. Today's high-density TCAMs consumes 12 to 15 Watts of power per chip, when the entire memory is enabled. This paper proposes an efficient representation of data in the forwarding tables and search algorithm which takes the algorithmic time complexity is O(loglog n). This paper also proposes an efficient way to reduce the index TCAM size by using y-Fast trie-partitioning algorithm.

Constructing scalable hierarchical switched openflow network using adaptive replacement of flow table management

Abstract: The need for building large networks is on the rise due to rapid growth of number of internet users and also in data centers where large number of switched networks are formed. Hierarchical structure is the most preferable way of handling complex systems of any kind. Hence in this paper we devise a large hierarchical switched network using OpenFlow and propose the concept of adaptive replacement in flow table to evict uncertain flows at a particular time. Open Flow is a protocol which separates the control plane and data plane which helps in managing flow table and also to place controllers in large hierarchical networks.

Efficient IP lookup using hybrid trie-based partitioning of TCAM-based open flow switches

Abstract: IP forwarding technique in open flow switch can be done by comparing the destination IP address, which is stored in forwarding table with the input IP prefix. Ternary content-addressable memory (TCAM) is one of the popular mechanisms to store and forward IP packet where flow entries are organized in sorted manner. Searching a prefix value in TCAM uses longest prefix match rather than exact match technique. The major drawback of TCAM is high power consumption (12---15 Watts per chip) due to increase in lookup time. The objective of this paper was to reduce the search time of a key, which is stored in the forwarding table. This paper also proposes an efficient way to represent data and to reduce the index TCAM size by using $$y$$y-fast trie-partitioning algorithm, and it will take search time complexity of $$O(loglog~n)$$O(loglogn).

SDN Flow Entry Management Using Reinforcement Learning

Abstract: Modern information technology services largely depend on cloud infrastructures to provide their services. These cloud infrastructures are built on top of Datacenter Networks (DCNs) constructed with high-speed links, fast switching gear, and redundancy to offer better flexibility and resiliency. In this environment, network traffic includes long-lived (elephant) and short-lived (mice) flows with partitioned/aggregated traffic patterns. Although SDN-based approaches can efficiently allocate networking resources for such flows, the overhead due to network reconfiguration can be significant. With limited capacity of Ternary Content-Addressable Memory (TCAM) deployed in an OpenFlow enabled switch, it is crucial to determine which forwarding rules should remain in the flow table and which rules should be processed by the SDN controller in case of a table-miss on the SDN switch. This is needed in order to obtain the flow entries that satisfy the goal of reducing the long-term control plane overhead introduced between the controller and the switches. To achieve this goal, we propose a machine learning technique that utilizes two variations of Reinforcement Learning (RL) algorithms—the first of which is a traditional RL-based algorithm, while the other is deep reinforcement learning-based. Emulation results using the RL algorithm show around 60% improvement in reducing the long-term control plane overhead and around 14% improvement in the table-hit ratio compared to the Multiple Bloom Filters (MBF) method, given a fixed size flow table of 4KB.

Advances in Software-Defined Technologies for Underwater Acoustic Sensor Networks: A Survey

Abstract: Underwater Acoustic Sensor Networks (UASNs) are an important technical means to explore the ocean realm. However, most UASNs rely on hardware infrastructures with poor flexibility and versatility. The systems typically deploy in a redundant manner, which not only leads to waste but also causes serious signal interference due to multiple noises in designated underwater regions. Software-Defined Networking (SDN) is a novel network paradigm, which provides an innovative approach to improve flexibility and reduce development risks greatly. Although SDN and UASNs are hot topics, there are currently few studies built on both. In this paper, we provide a comprehensive review on the advances in software-defined UASNs. First, we briefly present the background, and then we review the progress of the Software-Defined Radio (SDR), Cognitive Radio (CR), and SDN. Next, we introduce the current issues and potential research areas. Finally, we conclude the paper and present discussions. Based on this work, we hope to inspire more active studies and take a further step on software-defined UASNs with high performances.

Balanceamento entre segurança e desempenho na comunicação entre os planos de controle e dados em redes definidas por software

Abstract: Atualmente, a criptografia e utilizada como um padrao para proteger o trafego de dados pela Internet. Por exemplo, em uma rede definida por software, e possivel proteger o canal de comunicacao do plano de controle utilizando criptografia. Entretanto, essa funcionalidade pode aumentar o uso de recursos e consequentemente comprometer o desempenho de switches, principalmente ao considerar switches de baixo custo. Este trabalho visa a possibilidade de balancear seguranca e desempenho na comunicacao envolvendo controlador-switch em redes definidas por software. Para viabilizar essa possibilidade, a comunicacao e gerenciada levando em consideracao as mensagens conhecidas da API do Southbound para realizar um ataque. As mensagens que sao consideradas sensiveis sao gerenciadas com foco em seguranca e as mensagens atualmente consideradas inofensivas sao gerenciadas com foco em desempenho. Os resultados obtidos mostram que e possivel gerenciar com sucesso o balanceamento de seguranca e desempenho.

A dynamic adaptive timeout approach for SDN switch

Abstract: SDN have caught the attention of many researches and corporations because of its distinct advantages, but also exposed some problems meanwhile. Limited by today's technology, the size of flow table and the processing capacity of controller may lead to significant packet loss and network delay when matching fails. The matching failure means no corresponding entry in the flow table, which is mainly caused by inappropriate timeout. So we propose a dynamic adaptive approach in this paper for setting a proper timeout based on the actual network traffic, and it is supported by OpenFlow. At every sampling moment, the approach would execute the following steps. Firstly, estimate the remaining resource in the flow table. Then modify the timeout for the new coming flows to change their survival probability. And finally, choose an appropriate sampling period based on the network traffic. The results of experiments certified the effectiveness of the approach.

SDN Flow Entry Management Using Reinforcement Learning

Abstract: Modern information technology services largely depend on cloud infrastructures to provide their services. These cloud infrastructures are built on top of datacenter networks (DCNs) constructed with high-speed links, fast switching gear, and redundancy to offer better flexibility and resiliency. In this environment, network traffic includes long-lived (elephant) and short-lived (mice) flows with partitioned and aggregated traffic patterns. Although SDN-based approaches can efficiently allocate networking resources for such flows, the overhead due to network reconfiguration can be significant. With limited capacity of Ternary Content-Addressable Memory (TCAM) deployed in an OpenFlow enabled switch, it is crucial to determine which forwarding rules should remain in the flow table, and which rules should be processed by the SDN controller in case of a table-miss on the SDN switch. This is needed in order to obtain the flow entries that satisfy the goal of reducing the long-term control plane overhead introduced between the controller and the switches. To achieve this goal, we propose a machine learning technique that utilizes two variations of reinforcement learning (RL) algorithms-the first of which is traditional reinforcement learning algorithm based while the other is deep reinforcement learning based. Emulation results using the RL algorithm show around 60% improvement in reducing the long-term control plane overhead, and around 14% improvement in the table-hit ratio compared to the Multiple Bloom Filters (MBF) method given a fixed size flow table of 4KB.