Armando Caro

SCTP: a proposed standard for robust Internet data transport

The stream control transmission protocol (SCTP) is an evolving general purpose Internet transport protocol designed to bridge the gap between TCP and UDP. SCTP evolved from a telephony signaling protocol for IP networks and is now a proposed standard with the Internet Engineering Task Force. Like TCP, SCTP provides a reliable, full-duplex connection and mechanisms to control network congestion. However, SCTP expands transport layer possibilities beyond TCP and UDP, offering new delivery options that are particularly desirable for telephony signaling and multimedia applications.

SCTP in battlefield networks

The Stream Control Transmission Protocol (SCTP) is a new Internet standards track transport layer protocol SCTP was originally designed to transport PSTN signaling messages over IP networks, but is also capable of serving as a general purpose transport protocol. As such, SCTP provides an alternative that may be better able to satisfy the requirements of future battlefield networks than the traditional transport protocols, TCP and UDP. Unlike traditional transport protocols, SCTP allows multiple streams of messages within a single connection (or, in SCTP terminology, a single association). As the results in this paper show, this ability is particularly helpful in reducing latency for streaming multimedia in high loss environments. SCTP also provides features for multi-homing that may be helpful in high-mobility environments and additional security against denial-of-service attacks based on SYN flooding.

Using SCTP multihoming for fault tolerance and load balancing

Mission critical systems rely on redundancy at multiple levels to provide uninterrupted service during resource failures. Such systems when connected to IP networks often deliver network redundancy by multihoming their hosts. A host is multihomed if it can be addressed by multiple IP addresses. An endpoint’s IP address can become inaccessible, possibly due to an interface failure, severe congestion, or due to BGP’s slow route convergence around path outages. Redundancy at the network layer allows a host to be accessible even if one of its IP addresses becomes unreachable; packets can be rerouted to one of its alternate IP addresses.

Adaptive Dynamic Radio Open-source Intelligent Team (ADROIT): Cognitively-controlled Collaboration among SDR Nodes

The ADROIT project is building an open-source software-defined data radio, intended to be controlled by cognitive applications. The goal is to create a system that enables teams of radios, where each radio both has its own cognitive controls and the ability to collaborate with other radios, to create cognitive radio teams. The desire to create cognitive radio teams, and the goal of having an open-source system, requires a rich and carefully architected system that provides great flexibility (enabling cognitive applications to change the radios behavior) and also has a clear structure (both so that others may add or enhance the software, and also so that the system can be clearly modeled for cognitive applications). What follows is a summary of the ADROIT system and the key architectural features intended to enable cognitive radio teams.

Transport layer multihoming for fault tolerance in FCS networks

In this paper, we document a potential inefficiency in the current SCTP retransmission policy. The current scheme intends to improve the chance of success by exploiting the redundant paths between multihomed endpoints, but we have found that the current SCTP retransmission policy often degrades performance. We comparatively evaluate an alternative retransmission policy and show that the current SCTP retransmission policy unexpectedly performs worse under certain conditions. Our analysis exposes the problem and we discuss four possible solutions.

End-to-end failover thresholds for transport layer multihoming

SCTPs multihoming failure detection time depends on three tunable parameters: RTO.min (minimum retransmission timeout), RTO.max (maximum retransmission timeout), and Path.Max.Retrans (threshold number of consecutive timeouts that must be exceeded to detect failure). RFC2960 recommends Path.Max.Retrans = 5, which translates to a failure detection time of at least 63 seconds - unacceptable to many applications. This research investigates the tradeoff between a more aggressive (i.e., lower) threshold, and spurious failovers for the application of bulk file transfer. We surprisingly find that spurious failovers do not degrade overall performance, and sometimes actually improve goodput performance.

Rethinking End-to-End Failover with Transport Layer Multihoming

Using the application of bulk data transfer, we investigate end-to-end failover mechanisms and thresholds for transport protocols that support multihoming (e.g.,Sctp). First, we evaluate temporary failovers, and measure the tradeoff between aggressive (i.e., lower) thresholds and spurious failovers. We surprisingly find that spurious failovers do not degrade performance, and often actually improve goodput regardless of the paths ’ characteristics (bandwidth, delay, and loss rate). A permanent failover mechanism tries to avoid throttling the sending rate by not returning to a primary path when it recovers. We demonstrate that such a mechanism can be beneficial if the sender can estimate each path’s rtt and loss rate. We advocate a new approach to end-to-end failover that temporarily redirects traffic to an alternate path on the first sign of a potential failure (i.e., a timeout) on the primary path, but conservatively proceeds with failure detection of the primary path in the background.RésuméEn appliquant les transferts de données par rafale, nous étudions les mécanismes et les seuils de basculement de bout en bout dans les protocoles de transport (par exempleSctp) qui supportent la multidomiciliation. Nous évaluons tout d’abord des basculements temporaires et mesurons le compromis entre les seuils agressifs (les plus bas) et les faux basculements. Nous avons eu la surprise de découvrir que loin de dégrader les performances, les faux basculements améliorent souvent le débit utile et ce indépendamment des caractéristiques des chemins (bande passant, délai, et taux de perte). Un mécanisme de basculement permanent tente d’éviter de ralentir le taux d’émission en ne retournant pas à un chemin primaire lors d’un recouvrement d’erreur. Nous démontrons le bénéfice qu’apporte un tel mécanisme à un utilisateur capable d’estimer le temps de propagation en boucle (rit) et le taux de perte de chaque chemin. Nous préconisons une nouvelle approche de basculement de bout en bout qui redirige temporairement le trafic vers un chemin alternatif dès le premier signe d’une panne potentielle (expiration de temporisateur) sur le chemin primaire, mais traite de manière conservative et en arrière plan la détection de panne du chemin principal.

Testing environment for innovative transport protocols

This paper describes the development of a test environment for innovative transport protocols. Central to this work is the development of a universal transport library (UTL). The UTL is a library of transport protocols that provides application programmers the ability to write to a single application programming interface (API), then test their application with many different transport protocols. The UTL also allows for rapid prototyping of transport protocols at the user level. The UTL has been incorporated into two multimedia communication systems designed to provide better performance over lossy networks by using innovative transport protocol features: NETCICATS (a Network-Conscious Image Compression and Transmission System) and ReMDoR (a Remote Multimedia Document Retrieval system). These three tools facilitate the evaluation of flexible transport protocols and compression techniques for multimedia communications over lossy battlefield networks.

Network-conscious compressed image transmission over battlefield networks

We introduce an image compression and transmission system for battlefield networks. The system is based on network-conscious image compression, an approach to compression that does not simply maximize compression, but which optimizes overall performance when compressed images are transmitted over a lossy, packet-switched battlefield network. Using an application level framing philosophy, an image is compressed into path-maximum transmission unit-size application data units (ADUs) at the application layer. Each ADU is independent of others and carries its own semantics, that is, each ADU is a self-contained unit possessing all information necessary for decoding and displaying the information within that packet. Each independent ADU can be delivered to the receiving application out-of-order, thereby enabling faster progressive display of the image. We combine network-conscious image compression with an embedded focusing feature to provide a system that can be used in battlefield scenarios such as telemedicine or intelligence gathering.

The state of DTN evaluation

The most common definition of a challenged network, or Delay-Tolerant Network, is one in which there is no guarantee of a contemporaneous end-to-end path from source to destination. Because of the breadth of of the field, a wide variety of evaluation techniques have been developed and utilized. The goal of this panel is to showcase and discuss an interesting and current selection of evaluation efforts. In this year’s CHANTS submissions over a third used the ONE simulator from Aalto University [1, 2]. Keeping in mind that not all papers were based on simulation, it is fair to say that the ONE is becoming the de-facto simulation tool for certain types of DTN research. This is beneficial in that it facilitates comparisons between routing protocols, and allows researchers to focus their efforts on a single accessible tool. However the growing body of work that compares protocols that exist only in simulation may indicate an over-reliance on simulation and could lead to overlooking important practical technical challenges in the field. Another approach to evaluation is to deploy trial nodes in a real environment. Masato Tsuru has been involved in such field tests in collaboration with Japan’s NIICT and will be able to discuss their reasons for carrying out their experiments and the challenges involved. In addition the DARPA DTN project involved a considerable amount of field testing, and BBN’s Armando Caro will be able to give his perspective on the value and importance of their tests.