Jairo O. Esteban

Caesar: a content router for high speed forwarding

Internet users are interested in content regardless of its location; however, the current client/server architecture still requires requests to be directed to a specific server. Information-centric networking (ICN) is a recent vein that relaxes this requirement through the use of name-based forwarding, where forwarding decisions are based on content names instead of IP addresses. Despite previous name-based forwarding strategies have been proposed, almost none have actually built a content router. To fill this gap, in this paper we design and prototype a content router called Caesar for high-speed forwarding on content names. Caesar introduces several innovative features, including (i) a longest-prefix matching algorithm based on a novel data structure called prefix Bloom filter; (ii) an incremental design which allows for easy integration with existing protocols and network equipment; (iii) a forwarding scheme where multiple line cards collaborate in a distributed fashion; and (iv) support for offloading packet processing to graphics processing units (GPUs). We build Caesar as an enterprise router, and show that every line card sustains up to 10 Gbps using a forwarding table with more than 10 million content prefixes. Distributed forwarding allows the forwarding table to grow even further, and to scale linearly with the number of line cards at the cost of only a few microseconds in the packet processing latency. GPU offloading, in turn, trades off a few milliseconds of latency for a large speedup in the forwarding rate.

On SIP performance

The performance characteristics of Session Initiation Protocol (SIP) network elements determine what services can be supported by SIP networks. These elements must be able to provide customers with appropriate response times for new services, e.g., push-to-talk service. In IP-telephony converged networks, SIP-based signaling elements must meet the requirements of telephony signaling — that is, SIP-based servers will be required to process messages according to real-time constraints. Performance of SIP network elements also determines how much SIP networks cost. To support large customer bases, service providers must provide adequate performance on economical components. In this paper, we examine the main processing requirements of SIP elements and describe a number of performance metrics. We support our analyses with measurements of four different implementations of selected SIP functions.

Interactions between HTTP adaptive streaming and TCP

HTTP adaptive streaming (HAS) is quickly becoming a popular mechanism for delivering on-demand video content over the Internet. The chunked transmission and application-layer adaptation create a very different traffic pattern than traditional progressive video downloads where the entire video is downloaded with a single request. In this paper, we investigate experimentally the interplay between HAS and the network transport control protocol (TCP). We investigate the impact of network delay on achievable throughput and discover that HAS streams cannot fully utilize the available bandwidth due to the start and stop nature of HAS traffic patterns and its interaction with TCP. We investigate TCP pacing as a potential solution to this issue, particularly for packet losses that occur as a result of bursting packets into the network at the start of a transmission. We find that pacing can significantly increase a TCP flows congestion window but it does not necessarily translate into higher throughput. Instead, we find that packet losses at the end of chunk transmission have a greater impact on throughput.

Adaptive streaming: The network HAS to help

HTTP adaptive streaming (HAS) is becoming popular for video delivery because it dynamically provides high quality content. A HAS client does this by requesting a small chunk of video at a time, selecting the chunk quality based on inferred network conditions and its own buffer state. Because chunks are requested using Hypertext Transfer Protocol (HTTP), they traverse firewalls and are cacheable like any other Web content. Even though HAS is compatible with existing infrastructure, HASs near-real time demands and the new request patterns it causes require an understanding of the interaction between the network and HAS algorithms. In this paper, we evaluate the responsiveness of HAS algorithms under dynamic conditions and how it affects the overall user experience as well as key infrastructure resources, particularly intermediate caches. Relevant scenarios we evaluate include varying delays, available bandwidth, cache response times, and interaction with competing traffic.

Providing Local Content Discovery and Sharing in Mobile Tactical Networks

The tactical edge is filled with rich content, providing real-time intelligence and situation awareness to the warfighter. The current network architecture lacks support for lateral sharing of this critical content. Content is currently accessed from known servers and may require reach-back over bandwidth-constrained links. In this work we present SCALE, a Scalable Content-centric Architecture ensuring Locality and Efficiency. SCALE focuses on sharing data as it is created and is optimized for ad hoc collections of mobile nodes, rather than defined clients and servers. Any node can create and share content and nodes can discover content without knowing ahead of time where to look for it. SCALE provides mechanisms by which content is replicated throughout the network by organically caching content as it transits through the network or prepopulating content caches for existing content. SCALE provides a distributed index based on multi-node resolution, in which nodes dynamically take on responsibility for subsets of the content index based solely on being in a particular geographic location. We provide an evaluation of SCALEs mechanisms for providing local content discovery and sharing based on a SCALE prototype running on Android devices. These results show that SCALE can improve the probability that content is available locally and decrease the number of hops needed to access content. Dynamic ad hoc content sharing apps (e.g. social networking, microblogging, photo sharing, search, chat channels, etc.) can be built on top of SCALE to provide enhanced situation awareness and command and control operations for mobile tactical users.

WiLo: A Rate Determination Algorithm for HAS video in wireless networks and low-delay applications

In this paper, we describe WiLo, a Rate Determination Algorithm (RDA) for HTTP Adaptive Streaming (HAS) applications. We demonstrate that WiLo delivers stable video output and avoids buffer starvation, even in highly dynamic wireless networks where existing algorithms often struggle. We also show WiLo to be effective with very small buffers, making it useful for live streaming applications where delay is a concern.

Diabelli: An IMS simulation tool

The IP Multimedia Subsystem (IMS) standards, defined by the 3rd Generation Partnership Project (3GPP), specify a large number of functional units. The quantity and location of these units vary widely depending on network access technology, services, and market size. We have developed Diabelli, an IMS simulation tool that models IMS functional units. This tool allows network designers to specify over 80 different parameters to simulate the interaction of IMS elements.

ISE03-3: Towards Stateless Core: Improving SIP Proxy Scalability

The Third Generation Partnership Project (3GPP) specified the IP Multimedia Subsystem (IMS) standard. This standard uses the Session Initiation Protocol (SIP) to setup and teardown multimedia sessions through a potentially large chain of proxies. The IMS standard allows some core elements to be deployed as transaction stateful or stateless proxies. Moreover, SIP standard allows proxies decide when to store transaction information on a per request basis. In this paper, we examine the implementation of these core IMS proxies. We propose an algorithm that determines for each incoming request when to store transaction information. The goal of this algorithm is to increase throughput by reducing the number of transactions without compromising call setup times. Our algorithm uses information such as the incoming SIP request, CPU usage, and the number of retransmitted messages. We have implemented this algorithm in our IMS simulation tool. The results show that the system throughput can be improved by as much as 46%.

SCALE: a Content-Centric MANET

The authors present SCALE, a content-centric design for MANETs. When possible, SCALE favors routing based on content names to provide local and efficient content retrieval. Geographic hash table (GHT)-like resolution is used as a backup mechanism to ensure a target level of availability as well as scalability. This content retrieval routing decision is made by a mechanism that dynamically selects the best protocol while accounting for application needs. The author use content popularity as a fundamental criterion to switch between name-based routing and GHT-like resolution. Following a content-centric design, SCALE also supports on-path caching, where network nodes temporarily store copies of content for potential future retransmission on demand. However, we index data in the caches based on the content rather than the name of the content using fingerprinting: this strategy eliminates redundant data transmission when possible. Basic SCALE functionality had been implemented over a network of Galaxy Nexus phones running Android 4.1.1 which are connected in ad hoc mode. In this demonstration, publishing/requesting content by name, and returning content to the user with low latency, high availability, and little overhead in a dynamic wireless environment were demonstrated.