Content Centric Networking (CCN) and Named Data Networking (NDN) are popular ICN proposals that are widely accepted in the ICN community; however, they do not provide an efficient pub/sub mechanism. Hence, a content oriented pub/sub system named COPSS was developed to enhance the CCN/NDN protocols with efficient pub/sub capabilities. Internet houses powerful devices like routers and servers that can operate with the full-fledged implementation of such ICN protocols. However, Internet of Things (IoT) has become a growing topic of interest in recent years with billions of resource constrained devices expected to connect to the Internet in the near future. The current design to support IoT relies mainly on IP which has limited address space and hence cannot accommodate the increasing number of devices. Even though, IPv6 provides a large address space, IoT devices operate with constrained resources and hence, IPv6 protocol and its headers will induce additional overhead for their operation. Interestingly, we observed that IoTs are information centric in nature and therefore, ICN could be the more suitable candidate to support IoT environments. Although NDN and COPSS are designed for the Internet, their current full fledged implementations cannot be used by the resource constrained IoT devices. Therefore, CCN-lite was designed to provide a light weight, inter-operable version of the CCNx protocol to support the IoT devices. However, we show that communication in the IoT networks resemble pub/sub communication paradigm. However, CCN-lite like its ancestors (CCN/NDN) lacks the support for an efficient pub/sub mechanism while COPSS cannot be directly applied to the constrained IoT networks. Therefore, in this work, we develop COPSS-lite, an efficient and light weight implementation of pub/sub along with multi-hop routing to support the IoT networks. Essentially, COPSS-lite enhances CCN-lite with pub/sub capability with minimal overhead and further enables multi-hop connections by incorporating the famous RPL protocol for low power and lossy networks. Through evaluation using the real world sensor devices from the IoT Lab, we demonstrate the benefits of COPSS-lite in comparison with stand alone CCN-lite. Our results show that COPSS-lite is compact, operates on all platforms that support CCN-lite and significantly improves the performance of constrained devices in the IoT environments.

Ensuring high availability (HA) for software-based networks is a critical design feature that will help the adoption of software-based network functions (NFs) in production networks. It is important for NFs to avoid outages and maintain mission-critical operations. However, HA support for NFs on the critical data path can result in unacceptable performance degradation. We present REINFORCE, an integrated framework to support efficient resiliency for NF service chains. REINFORCE includes timely failure detection and consistent failover mechanisms. REINFORCE replicates state to standby NFs (local and remote) while enforcing correctness. It minimizes the number of state transfers by exploiting the concept of external synchrony, and leverages opportunistic batching and multi-buffering to optimize performance. Experimental results show that, even at line-rate packet processing (10 Gbps), REINFORCE achieves chain-level failover across servers in a LAN within 10ms, incurring less than 10% performance overhead, and adds average latency only ~400 μs, with a worst-case latency of less than 1ms. REINFORCE also recovers from software failures within the same node in less than 100 μs, incurring less than 1% performance overhead and adds less than 5 μs latency during normal operation.

Managing Network Function (NF) service chains requires careful system resource management. We propose NFVnice, a user space NF scheduling and service chain management framework to provide fair, efficient and dynamic resource scheduling capabilities on Network Function Virtualization (NFV) platforms. The NFVnice framework monitors load on a service chain at high frequency (1000Hz) and employs backpressure to shed load early in the service chain, thereby preventing wasted work. Borrowing concepts such as rate proportional scheduling from hardware packet schedulers, CPU shares are computed by accounting for heterogeneous packet processing costs of NFs, I/O, and traffic arrival characteristics. By leveraging cgroups, a user space process scheduling abstraction exposed by the operating system, NFVnice is capable of controlling when network functions should be scheduled. NFVnice improves NF performance by complementing the capabilities of the OS scheduler but without requiring changes to the OS's scheduling mechanisms. Our controlled experiments show that NFVnice provides the appropriate rate-cost proportional fair share of CPU to NFs and significantly improves NF performance (throughput and latency) by reducing wasted work across an NF chain, compared to using the default OS scheduler. NFVnice achieves this even for heterogeneous NFs with vastly different computational costs and for heterogeneous workloads.

The currently trending paradigms of edge and fog computing attempt to provide services close to the end user, to meet the demands of latency-sensitive applications and to limit bandwidth consumption in the network core. One open issue is the pricing of edge and fog resources. Current pricing schemes are usually oligopolistic and not fair. In this work, we propose DMC, a marketplace that can dynamically determine the fair price for arbitrary resource types and instances based on supply and demand existing at that period. Unlike the state-of-the-art solutions, DMC performs integral allocation of resources and thereby avoids the unbounded integrality gap. Additionally, DMC provides differential pricing among instances to allow varying prices based on the perceived value of a resource. We evaluate DMC with both heavy and non-heavy tailed distributions to reflect diverse buying interests and the number of resources sold to demonstrate the feasibility of our solution for several realistic scenarios. We observe that (i) DMC arrives at market-clearing prices; (ii) DMC generates 10x to 100x more profit than state-of-the-art solutions, while still maximizing the Nash Social Welfare to achieve prices that are fair to both buyers and the resource providers; and (iii) the computation time for DMC does not exceed 10 seconds even in the case of 500 resource types with 500 buyers each, making it applicable for real-time use cases.

Currently, cloud computing is a primary enabler of new paradigms such as edge and fog computing. One open issue is the pricing of services or resources. Current pricing schemes are usually oligopolistic and not fair. In this work, we propose OFM, an online learning based marketplace that dynamically determines the price for arbitrary resource types based on supply and demand existing at that period. Unlike state of the art solutions, OFM can handle an arbitrary number of customers and resource types at every instance of time. It further performs integral allocation of resources and thereby avoids the unbounded integrality gap. We evaluate OFM with both real and synthetic datasets to reflect varying buying interests, the number of resources sold and market volatility to demonstrate the feasibility of our solution for several realistic scenarios. We observe that (i) OFM achieves about 9% of optimal prices and maximizes the Nash social welfare (NSW); (ii) OFM converges faster and works with different data distributions; and (iii) OFM scales for a large number of resources and buyers and computational time is in the order of microseconds, making it applicable for real-time use cases especially in edge markets.

Ensuring high availability (HA) for software-based networks is a critical design feature that will help the adoption of software-based network functions (NFs) in production networks. It is important for NFs to avoid outages and maintain mission-critical operations. However, HA support for NFs on the critical data path can result in unacceptable performance degradation. We present REINFORCE, an integrated framework to support efficient resiliency for NFs and NF service chains. REINFORCE includes timely failure detection and consistent failover mechanisms. REINFORCE replicates state to standby NFs (local and remote) while enforcing correctness. It minimizes the number of state transfers by exploiting the concept of external synchrony, and leverages opportunistic batching and multi-buffering to optimize performance. Experimental results show that, even at line-rate packet processing (10 Gbps), REINFORCE achieves chain-level failover across servers in a LAN (or within the same node) within 10ms (100/μs), incurring less than 10% (1%) performance overhead, and adds average latency of only ~400/μs (5/μs), with a worst-case latency of less than 1ms (10/μs).

Internet of Things (IoT) is a growing topic of interest. Billions of IoT devices are expected to connect to the Internet in the near future. These devices differ from the traditional devices operated in the Internet. In this paper, we argue that an information-centric networking (ICN), a new networking paradigm, is a more suitable architecture for the IoT compared to the currently prevailing IP-based network. We observe that recent works that propose to use ICN for IoT, either do not cover the need to integrate sensor networks with the Internet to realize IoT or do so inefficiently. There is a need to understand effective ways to integrate the various heterogeneous sensor networks with the Internet without affecting their current mode of operation. In this paper, we study the essential requirements for integrating sensor networks to the Internet. We provide an architecture with gateways for paving a way for the sensor networks to become a part of the IoT family. We further provide a naming schema for efficient operation of the resource constrained sensor networks, discuss mobility, security, communication patterns, and propose the most suitable choices for IoT networks.

Managing Network Function (NF) service chains requires careful system resource management. We propose NFVnice, a user space NF scheduling and service chain management framework to provide fair, efficient and dynamic resource scheduling capabilities on Network Function Virtualization (NFV) platforms. The NFVnice framework monitors load on a service chain at high frequency (1000Hz) and employs backpressure to shed load early in the service chain, thereby preventing wasted work. Borrowing concepts such as rate proportional scheduling from hardware packet schedulers, CPU shares are computed by accounting for heterogeneous packet processing costs of NFs, I/O, and traffic arrival characteristics. By leveraging cgroups, a user space process scheduling abstraction exposed by the operating system, NFVnice is capable of controlling when network functions should be scheduled. NFVnice improves NF performance by complementing the capabilities of the OS scheduler but without requiring changes to the OS's scheduling mechanisms. Our controlled experiments show that NFVnice provides the appropriate rate-cost proportional fair share of CPU to NFs and significantly improves NF performance (throughput and loss) by reducing wasted work across an NF chain, compared to using the default OS scheduler. NFVnice achieves this even for heterogeneous NFs with vastly different computational costs and for heterogeneous workloads.

The vast majority of mobile data transfers today follow the traditional client-server model. Although in the fixed network P2P approaches have been exploited and shown to be very efficient, in the mobile domain there has been limited attempt to leverage on P2P (D2D) for large-scale content distribution (i.e., not DTN-like, point-to-point message transfers). In this paper, we explore the potential of a user-operated, smartphone-centric content distribution model for smartphone applications. In particular, we assume source nodes that are updated directly from the content provider (e.g., BBC, CNN), whenever updates are available; destination nodes are then directly updated by source nodes in a D2D manner. We leverage on sophisticated information-aware and application-centric connectivity techniques to distribute content between mobile devices in densely-populated urban environments. Our target is to investigate the feasibility of an opportunistic content distribution network in an attempt to achieve widespread distribution of heavy content (e.g., video files) to the majority of the destination nodes. We propose ubiCDN as a ubiquitous, user-operated and distributed CDN for mobile applications.

Information-Centric Networking (ICN) treats content as a first-class entity - each content has a unique identity and ICN routers forward traffic based on content identity rather than the locations of the content. This provides benefits like dynamic request routing, caching and mobility support. The choice of naming schema (flat vs. hierarchical) is a fundamental design choice in ICN which determines the functional separation between the network layer and the application layer. With hierarchical names, the network layer is cognizant of the semantics of hierarchical names. Name space management is also part of network layer. ICN architectures using flat names leave these to the application layer. The naming schema affects the performance and scalability of the network in terms of forwarding efficiency, routing table size and name space size. This paper provides both qualitative and quantitative comparison on the two naming schemas using these metrics, noting that they are interdependent. We seek to understand which naming schema would be better for a high-performance, scalable ICN architecture.

In the aftermath of natural disasters (e.g., earthquakes and hurricanes), information dissemination to and from citizens and authorities often involves the use of mobile network services. They may do so to seek help, distribute critical information, and possibly to confirm the safety of relatives and friends. The traffic on the cellular network typically increases substantially in the aftermath of such disasters. The situation is further compounded by disruptions caused by damage to the network infrastructure and the non-availability of power. Networks are fragmented, and communication is intermittent and disruption prone. In this paper, we propose a name-based many-to-many communication service for such fragmented networks. The key feature of this service is an integration of several communication methods, such as push-based and pull-based. The service effectively supports a variety of network situations by leveraging the features of information-centric networks.

Information Centric Networks (ICN) enable accessing data oblivious of its location, by allowing end-systems to retrieve content based on names. But, architectures such as Named Data Networking (NDN) and Content Oriented Publish/Subscribe System (COPSS) do not yet provide a mechanism for end-system applications to obtain these names. There is a need for an object resolution system that addresses a most important and as yet unimplemented component of obtaining a name in ICN. In this paper, we propose ORICE, an architectural design for Object Resolution services in Information-Centric Environment that satisfies this need. The architecture enables intelligent resolution service by placing the service in the application layer and allows for the service diversity by separating the name space management from resolution service. Through preliminary evaluation, we show that with the help of ORICE, the states stored in the network can be dramatically reduced while ensuring complete data delivery.

Disasters result in disruptions to the communication infrastructure, typically forming a multitude of fragmented networks. This makes it difficult to exchange even simple while critical messages with large portions of the affected population, including first-responders and government authorities. In this paper, we focus on a content driven data retrieval model and propose an enhanced information-centric network (ICN) approach to provide communication resilience to such disruption-prone, delay-tolerant networks. The message exchange between communities and mules takes into account the limited available resources and the fact that the interest and the data might traverse completely different paths, unlike the assumption in many of the existing ICN solutions. Moreover, we argue for the separation of the the logical faces from the actual physical interfaces taking into account the fact that the data mules behave as mobile routers and are connected to different nodes at different time periods, unlike the assumptions made in existing ICN solutions. Our preliminary evaluations show that our solution is able to outperform other approaches such as “Epidemic” and “SprayAndWait” with respect to the latency, response probability and overall performance while improving the performance (latency and response probability) for high priority messages.

Information Centric Networking (ICN), a novel network paradigm, places the focus on the content instead of the end-hosts. ICN addresses content by names instead of locations and can ease content retrieval and improve network efficiency. Ongoing work attempts to extend the use of ICN to scenarios such as real time communication, group communication, push services, and addresses the issue of migration from the current network and coexistence of different network paradigms. In this work, we argue that by extending the current design of ICN from a “host-to-name” to a “name-to-name” architecture, the utility and efficiency of ICN could be further increased. We propose Internames, an architectural framework in which names are used to identify all entities involved in communication: content, users, devices, logical points, and services. Internames is envisioned to be an overarching name-to-name communication primitive that is fully compatible with ICN principles, accommodates the coexistence (or gradual migration) of different network realms (e.g., IP, ICN, VANET) and is suitable for application scenarios where ICN is somehow limited by its reliance on a “host-to-name” approach. In this paper, we provide early insights into the Internames architecture by leveraging on the work done by the research community and identify components and challenges that require more detailed investigation.

Many network operators have recently started to deploy their own Content Delivery Networks (CDNs) within their administrative domain. Having a CDN deployed within an operator's network has the advantage of potentially placing caches very close to the end user, thereby increasing Quality of Service for the user, while at the same time saving delivery costs for the operator. However, if each network operator hosts its own CDN, there is a need for interconnection among individual CDNs. The standardization of such CDN Interconnection (CDNI) hence has recently started in the IETF and ETSI. In this context, we present the first prototype implementation and evaluation of CDNI request routing facilitated by means of Application Layer Traffic Optimization (ALTO). By advertising network layer topology costs as well as conveying aggregated load on cache clusters, ALTO can enable an upstream CDN to select the best downstream CDN for each content request it intends to redirect. Our experiments study how — by using ALTO — upstream CDN can achieve cost savings while also taking the overall load of the downstream CDN into account.

This work demonstrates the benefit brought by COPSS for an efficient notification service, including the convenience of hierarchical group management, network efficiency and timeliness in delivering the notification. The work will also demonstrate a simple first-step authorization. Future work will include a more complete authorization, authentication, encryption and mobility support for a scalable notification service.

Thin-client based solutions allow users to connect to remote servers and access content that is running on these servers within a virtual PC. With the advent of cloud based solutions, thin-client deployments running on remote Data-centers are increasingly popular. Unfortunately, since the traffic has to traverse through the Internet, issues such as latency, packet drops, and potentially congestion are introduced, which in turn affect the Quality of Experience (QoE) for the user. Our objective is to provide preferential treatment to certain thin-client flows over the rest of the thin-client flows traversing the same intermediate node, based on the application that the user is using at a given point in time. This is not straightforward, as it is not easy to identify what actual application is running inside a given thin client session, given that thin client protocols essentially only send bitmaps of the desktop to the client and in addition are often encrypting traffic. Assuming that some statistical mechanisms for application identification for thin client connections exists, the challenge we address is how to exploit this information for QoE-driven scheduling. We present a scheme that allows the prioritization of thin client flows that are serving delay-sensitive applications, i.e., prioritizing flows based on the specific QoE requirements of the dynamically changing individual applications running within persistent thin client traffic flows. Our solution is essentially a hybrid scheduling scheme that takes into account the dynamically changing delay and bandwidth requirements of inner (i.e., tunneled in the thin client protocol) applications to prioritize flows that are close to an application-dependent QoE threshold. Our evaluation based on a prototype implementation reveals that our algorithm is indeed effective in dynamically prioritizing persistent thin-client flows based on the dynamically changing inner applications running within the flows.

Content-Centric Networking (CCN) seeks to meet the content-centric needs of users. In this paper, we propose hybrid-COPSS, a hybrid content-centric architecture. We build on the previously proposed Content-Oriented Publish/Subscribe System (COPSS) to address incremental deployment of CCN and elegantly combine the functionality of content-centric networks with the efficiency of IP-based forwarding including IP multicast. Furthermore, we propose an approach for incremental deployment of caches in generic query/response CCN environments that optimizes latency and network load. To overcome the lack of inter-domain IP multicast, hybrid-COPSS uses COPSS multicast with shortcuts in the CCN overlay. Our hybrid approach would also be applicable to the Named Data Networking framework.

Information-Centric Networking provides substantial flexibility for users to obtain information without knowing the source of information or its current location. With users increasingly focused on an online world, an emerging challenge for the network infrastructure is to support Massively Multiplayer Online Role Playing Game (MMORPG). Currently, MMORPG is built on IP infrastructure with the primary responsibility resting on servers for disseminating control messages and predicting/retrieving objects belonging to each player's view. Scale and timeliness are major challenges of such a server-oriented gaming architecture. Limited server resources significantly impair the user's interactive experience, requiring game implementations to limit the number of players in a single game instance. We propose Gaming over COPSS (G-COPSS), a distributed communication infrastructure using a Content-Oriented Pub/Sub System (COPSS) to enable efficient decentralized information dissemination in MMORPG, jointly exploiting the network and end-systems for player management and information dissemination. G-COPSS aims to scale well in the number of players in a single game, while still meeting users' response time requirements. We have implemented G-COPSS on top of the open-source CCNx implementation. We use a simple game with a hierarchical map to carefully micro benchmark the implementation and the processing involved in managing game dynamics. We have also micro benchmarked the game based on NDN and a server with an IP infrastructure. We emulate an application that is particularly emblematic of MMORPG -- Counter-Strike -- but one in which all players share a hierarchical structured map. Using trace-driven simulation, we demonstrate that G-COPSS can achieve high scalability and tight timeliness requirements of MMORPG. The simulator is parameterized based on micro benchmarks of our implementation. Our evaluations show that G-COPSS provides orders of magnitude improvement in update latency and a factor of two reduction in aggregate network load compared to a server-based implementation.

Content-Centric Networks (CCN) provide substantial flexibility for users to obtain information without regard to the source of the information or its current location. Publish/subscribe (pub/sub) systems have gained popularity in society to provide the convenience of removing the temporal dependency of the user having to indicate an interest each time he or she wants to receive a particular piece of related information. Currently, on the Internet, such pub/sub systems have been built on top of an IP-based network with the additional responsibility placed on the end-systems and servers to do the work of getting a piece of information to interested recipients. We propose Content-Oriented Pub/Sub System (COPSS) to achieve an efficient pub/sub capability for CCN. COPSS enhances the heretofore inherently pull-based CCN architectures proposed by integrating a push based multicast capability at the content-centric layer. We emulate an application that is particularly emblematic of a pub/sub environment - Twitter - but one where subscribers are interested in content (e.g., identified by keywords), rather than tweets from a particular individual. Using trace-driven simulation, we demonstrate that our architecture can achieve a scalable and efficient content centric pub/sub network. The simulator is parameterized using the results of careful micro benchmarking of the open source CCN implementation and of standard IP based forwarding. Our evaluations show that COPSS provides considerable performance improvements in terms of aggregate network load, publisher load and subscriber experience compared to that of a traditional IP infrastructure.

Supporting Massively Multiplayer Online Role Playing Games (MMORPG) is a significant challenge. MMORPGs have become very popular because of their attractive structuring and incorporation of realistic and creative scenarios. World of Warcraft and Counter-Strike are examples of such games and are characterized by high interactivity (need very low network latency). Every action an individual player performs typically needs to be communicated to all the related players and the players need to react according to the 'current' environment and the cumulative actions of all the players. Games like Second Life involve a large number of players and require a persistent view of the world that is usually managed by a dedicated server (e.g., one that is hosted by the game's publisher). The load on such a server for player management and communication can be significant, and is likely to be a source of substantial latency.

With users increasingly focused on an online world, an emerging challenge for the network infrastructure is the need to support Massively Multiplayer Online Role Playing Games (MMORPG). This is an application domain that is attracting more players than ever before, very often with players distributed over a metropolitan area. Currently, MMORPG are built on an IP infrastructure with the primary responsibility on servers to do the work of disseminating control messages and having to predict/retrieve objects in each player's view. Limited server resources significantly impair the user's interactive experience. Modern fast-paced action games that run on a client/server architecture limit the number of players who can interact simultaneously since the server needs to handle the frequent updates and disseminate them. Scale and timeliness are major challenges of such a server-oriented gaming architecture. We propose Gaming over COPSS (G-COPSS), a communication infrastructure using a Content-Oriented Pub/Sub System (COPSS) to enable efficient decentralized information dissemination in MMORPG, exploiting the network and the end-systems for player management and information dissemination. We emulate an application that is particularly emblematic of MMORPG - Counter-Strike - but one in which all the players share a hierarchical structured map. Using trace-driven simulation, we demonstrate that G-COPSS can achieve high scalability and tight timeliness requirements of MMORPG. The simulator is parameterized using the results of careful microbenchmarking of the open-source CCN implementation and of standard IP-based forwarding. Our evaluations show that G-COPSS provides considerable performance improvement in terms of aggregate network load and update latency compared to that of a traditional IP server-based infrastructure.

Delay-insensitive applications such as P2P file sharing, data center backups and software updates generate substantial amounts of traffic. This traffic, transported potentially over multiple TCP connections, competes with traffic from other possibly interactive applications. Today, with TCP, they compete on an equal level for each individual connection. In this paper, we propose a new TCP variant for such delay-insensitive applications, which we call Network Friendly TCP (NF-TCP). NF-TCP is responsive to available bandwidth, seeking to quickly and efficiently utilize the same in a congestion-free situation, while backing-off more aggressively than standard TCP on encountering competing traffic in a congested network. NF-TCP uses a novel combination of utilizing measurement of available bandwidth and ECN-based congestion avoidance techniques to ensure that it is truly friendly to existing TCP connections. We evaluate the performance of NF-TCP through ns-2 simulations and present the initial results enlightening the friendly nature of NF-TCP compared to standard TCP.

Recent trends in applications require per-flow treatment instead of the packet-based treatment inherent in the present day Internet. This requirement is analogous to trying to make a phone call in a standard PSTN line or cellular network and either receiving a busy tone or call accept. This paper explores the potential of providing flow-based admission control and termination functionality to an MPLS-based DiffServ network by applying pre-congestion notification (PCN). MPLS-based DiffServ networks are state of the art technologies used in many carrier backbones to provide quality of service for different traffic classes. The standard PCN scheme as developed by the IETF requires three code-points, one of which finds application in differentiating between PCN and non-PCN traffic, while the other two are used for PCN congestion marking to indicate admission control and flow termination, respectively. An MPLS-based DiffServ network on the other hand is limited as only two code-points are available. To overcome this limitation, we propose various schemes, namely: 1) one-codepoint signaling for admission control (AC), 2) one-codepoint signaling for flow termination (FT), and 3) one-codepoint signaling for 3 states (no congestion, AC and FT). The three schemes use the available code-point for different purposes and require different behaviors from the involved PCN nodes. We evaluate them using simulations. The results suggest that the third approach is able to effectively signal the three states without incurring much inaccuracy. The success rate with respect to the quality of experience (QoE) of existing flows, error recovery and throughput is not compromised either. To the best of our knowledge, this is the first known attempt to study the use of PCN in an MPLS-based DiffServ network.

This paper presents an evaluation of the quality of Service Signalling Layer Protocol (QoS-NSLP) of the NSIS (next steps in signalling) protocol suite. The QoS-NSLP in combination with the NSIS Transport Layer Protocol (NTLP) is proposed by the Internet Engineering Task Force (IETF) as an alternative to the Resource reSerVation Protocol (RSVP). We describe our implementations of the software architecture, both on a network simulator and on a Linux implementation. Both implementations are used in a complimentary manner to illustrate the performance of the QoS-NSLP protocol. The results show the performance of QoS-NSLP with respect to resource consumption, packet processing time, session set up time, refresh interval and protocol overhead. Furthermore, we analyse the protocol performance during route change scenarios.