As the computational capabilities of a supercomputer transition from petaflops to exaflops, more compute processes work concurrently to accomplish tasks, requiring more communication. This results in using an increasing number of software and hardware components, which in turn, increases the probability of abnormal events and failures. We present a solution that improves resilience against transient events in network communication. We observe that the coupling of the session and transport semantics in implementations inhibits recovery from transient failures. Our proposal, a session-layer intermediary (SLIM), serves as a shim layer on top of the interconnect's interface and enables separation of session and transport semantics. We use Open MPI as a case study where SLIM exposes an interface to the Byte Transfer Layer framework. This approach manages transient faults with the underlying transport, by trapping and resolving them and thus not allowing them to cascade into failed MPI primitives. Preliminary results show that the introduction of SLIM delivers resilience and does so without incurring any performance impact, either in latency or throughput. In future, we plan to include other interconnects, such as OpenIB, and enable tolerance for transient network failures.

Increasingly, communication requires more from the network stack, e.g., seamless handoff and synchronization of state between multiple participants. Due to the lack of support for desired functionality, networking libraries are created to fill the void. This leads to considerable duplication of effort and complicates cross-platform development. Furthermore, the means for extending legacy protocol stacks is largely exhausted (e.g., the TCP options space in the SYN message is mostly allocated), making the addition of future extensions much more challenging. In this paper, we tease apart elements of session management that are currently conflated with the transport semantics in TCP and highlight the need for sessions in contemporary communications. Next, we propose session, flow, and endpoint abstractions that lead to a clearer description of advanced communication models. This effort results in an extensible session-layer intermediary (SLIM) that leverages the above abstractions to support the additional functionality needed by modern applications, such as mobility, communication between two or more participants, and dynamic reconfiguration. SLIM's approach also provides the means for future extensibility of the network stack in a backward-compatible way, thus enabling incremental adoption.

Increasingly, communication requires more from the network stack. Due to missing functionality, we see a proliferation of networking libraries that attempt to fill the void (e.g., iOS to OSX Handoff and Google Cast SDK). This leads to considerable duplication of effort. Further, the provisions for extending legacy protocol stacks is largely exhausted (e.g., TCP options space is mostly allocated) making the addition of future extensions much more challenging. We present SLIM, an extensible session-layer intermediary that extracts the duplicate functionality from modern networking libraries and provides the means for future extensibility to the network stack. SLIM enables mobility, multiparty communication, and dynamic reconfiguration of the network stack in a straightforward and elegant way. SLIM includes an out-of-band signaling channel, which not only enables reconfiguration, but also allows for incremental evolution of the stack. To start, we tease out elements of session management which are currently conflated with transport semantics in TCP. Doing so highlights the need for sessions in contemporary use cases. Next, we propose session, flow and endpoint abstractions that allow application developers to describe communication between any number of participants. The abstractions apply to individual or a group communication allowing them to be managed as one. We describe the abstractions and evaluate them in terms of typical communication patterns. We demonstrate the abstractions via a prototype implementation of SLIM.

The bandwidth utilization in traditional TCP protocols (e.g., TCP New Reno) suffers over high-latency and high-bandwidth links due to the inherent characteristics of TCP congestion control. Conventional methods of improving throughput cannot be applied per se for streaming applications. The challenge is exacerbated by “big data” applications such as with the Long Wavelength Array data that is generated at a rate of up to 4 terabytes per hour. To improve bandwidth utilization, we introduce layer-4 relay(s) that enable the pipelining of TCP connections. That is, a traditional end-to-end connection is split into independent streams, each with shorter latencies, that are then concatenated (or cascaded) together to form an equivalent end-to-end TCP connection. This addresses the root cause by decreasing the latency over which the congestion-control protocol operates. To understand when relays are beneficial, we present an analytical model, empirical data and its analyses, to validate our argument and to characterize the impact of latency and available bandwidth on throughput. We also provide insight into how relays may be setup to achieve better bandwidth utilization.

Current technologies that support live migration require that the virtual machine (VM) retain its IP network address. As a consequence, VM migration is oftentimes restricted to movement within an IP subnet or entails interrupted network connectivity to allow the VM to migrate. Thus, migrating VMs beyond subnets becomes a significant challenge for the purposes of load balancing, moving computation close to data sources, or connectivity recovery during natural disasters. Conventional approaches use tunneling, routing, and layer-2 expansion methods to extend the network to geographically disparate locations, thereby transforming the problem of migration between subnets to migration within a subnet. These approaches, however, increase complexity and involve considerable human involvement. The contribution of our paper is to address the aforementioned shortcomings by enabling VM migration across subnets and doing so with uninterrupted network connectivity. We make the case that decoupling IP addresses from the notion of transport endpoints is the key to solving a host of problems, including seamless VM migration and mobility. We demonstrate that VMs can be migrated seamlessly between different subnets - without losing network state - by presenting a backward-compatible prototype implementation and a case study.

Saturating high capacity and high latency paths is a challenge with vanilla TCP implementations. This is primarily due to congestion-control algorithms which adapt window sizes when acknowledgements are received. With large latencies, the congestion-control algorithms have to wait longer to respond to network conditions (e.g., congestion), and thus result in less aggregate throughput. We argue that throughput can be improved if we reduce the impact of large end-to-end latencies by introducing layer-4 relays along the path. Such relays would enable a cascade of TCP connections, each with lower latency, resulting in better aggregate throughput. This would directly benefit typical applications as well as BIG DATA applications in distributed HPC. We present empirical results supporting our hypothesis.

The philosophy upon which the Internet was built places the intelligence close to the edge. As the Internet has matured, intermediate devices or middleboxes, such as firewalls or application gateways, have been introduced, thereby weakening the end-to-end nature of the network. As a result, applications must often modify their behavior to accommodate the middleboxes. This is is especially true in the case of transient failure of stateful devices. The failure of a middlebox causes it to lose the state it maintained, causing the failure of the associated TCP connections. Rather than assign the responsibility for recovery to applications, we incorporate a mechanism called an isolation boundary into TCP itself. The isolation boundary maintains a small amount of state across TCP connections, thus enabling reconnection. Furthermore, it does so without breaking backward compatibility with existing TCP. We present an implementation of the isolation boundary in the FreeBSD kernel and demonstrate its backward compatibility with TCP. We quantify the performance impact of the proposed mechanism on the establishment of new and resumed connections for both legacy and extended TCP stacks.

Many algorithms have been proposed in the last decade to detect traffic anomalies in enterprise networks. However, most of these algorithms cannot detect anomalies that occur beyond enterprise boundaries. Anomaly monitoring and detection on end-to-end Internet paths, although important for network operations, is challenging due to lack of access and control over intermediate network devices. In this paper, we propose an algorithm that detects anomalies or significant events on an end-to-end Internet path by monitoring the path's available bandwidth. We first evaluate existing algorithms on a comprehensive dataset of more than a million bandwidth measurements spanning three years. We show that existing algorithms do not incorporate the typical behavior of a path in the anomaly detection process and consequently incur accuracy degradations. We therefore propose to filter noisy bandwidth measurements to extract a typical or baseline statistical distribution of a path's bandwidth. This baseline model is in turn leveraged in a generic decision-theoretic framework to provide timely detection of significant path events. We show that the proposed detector provides highly accurate performance and easily surpasses the accuracy of existing techniques.

Geolocation is a leading Internet trend due to its significance in the commercial and services realm. So far, the network infrastructural dependency of IP geolocation has restricted its use to North America and Europe. However, the real potential of geolocation is in being a global rather than just a regional application. Since the last decade world-wide Internet usage and performance have increased exponentially. This directly urges the need to re-evaluate the performance of geolocation in more of the world. In this paper we first analyze various parameters in different regions. These are parameters that we believe affect geolocation performance and include: average distance and delay between landmarks and target, density of landmarks around a target, correlation between delay and distance around a target and the stability of proportionality between round trip times (RTT) and distance (referred to as alpha, α) around the target. We then test the performance of several well known geolocation techniques in different regions. These techniques include: Shortest Ping, TULIP, Constraint Based Geolocation with Speed of Internet, Constraint Based Geolocation and Topology Based Geolocation. We then quantify the estimates of each technique in relation to the values of the aforementioned parameters. Finally we analyze the impact of selecting a geometric technique (trilateration, multilateration and apollonius) on the performance of a geolocation solution.

Many algorithms have been proposed in the last decade to detect traffic anomalies in enterprise networks. However, most of these algorithms cannot detect anomalies that occur beyond enterprise boundaries. Performance monitoring and anomaly detection on end-to-end Internet paths, although important for network operations, is challenging due to lack of access and control over intermediate network devices. In this paper, we propose an algorithm that detects anomalies or significant events on an end-to-end Internet path by monitoring the path’s available bandwidth. We first evaluate existing algorithms on a comprehensive dataset of more than a million bandwidth measurements spanning three years. We show that incorporating the typical behaviour of a path in the process of anomaly detection improves accuracy. We therefore propose to filter noisy bandwidth measurements to extract the typical behaviour or baseline statistical distribution of a path’s bandwidth. This baseline model is in turn leveraged in a generic decision-theoretic framework to provide timely detection of significant path events. We show that the proposed detector provides high accuracy and surpasses the accuracy of existing techniques.

The growth of the Internet has ushered in and established the “Information Age.” However, its success has also arguably increased the difficulty of incorporating innovative changes that are needed to develop further functionality for next-generation networked applications. From the transport perspective, the desired functionality includes (1) supporting multiple network paths, (2) providing transport over hybrid networks (e.g., using both packetand circuit-switched networks), and (3) decoupling upper-layer services from endpoint-naming semantics. The need for functionality — such as transport composability — has been reiterated in recent research and leads to an apparent dilemma: TCP, the ubiquitous transport protocol, neither admits such functionality in its present form nor does it seem possible to add it without substantial modifications. Furthermore, radical changes — whether through incompatible extensions or by creating a completely new protocol — will not be easily accepted. In contrast to the apparent dilemma, we argue that a backward-compatible modification to TCP that supports increased functionality is possible without incurring significant burden in additional protocol exchange. The lightweight mechanism, built upon a set of TCP options, establishes an isolation boundary between TCP and the application. The boundary separates an application data stream from the TCP transport flow. Further, it provides for the establishment of a control channel that allows additional capabilities to be negotiated dynamically throughout the lifetime of the communication. In short, the mechanism provides a simple “hook” into TCP with which new features can be realized. This increases the freedom to evolve TCP while maintaining compatibility, thereby facilitating incremental adoption.

The complex software development scenarios for mobile/ hand-held devices operating in wireless environments require adaptation to the variations in the environment (such as fluctuating bandwidth). This translates to maintenance of service availability in preferably all circumstances. In this paper we propose that a mobile computing system (for hand-held, wireless devices) must be based on the combination of reflection, remote evaluation and code mobility mechanisms such that the communication framework allows developers to design disconnection-aware applications which maintain service availability in case of varying circumstances by automatically redeploying essential components to appropriate locations. This not only allows the application to continue executing in varying conditions, but also in entirely disconnected modes.

By way of introduction this talk will briefly illustrate the desperate straights of the African Internet related Infras- tructure. It will then go on to justify why and how we measure Internet Performance for countries of the world containing over 98% of the world's population. Using these measurements it will illustrate the overall Internet performance for the world in particular throughput, losses, Round Trip Times, jitter, connectivity etc., Particular emphasis will be paid to the performance for over 50 African countries and the differences between various regions and countries of Africa. It will demonstrate the performance trends for the last decade, in particular illustrating how Africa is not only behind all other regions and one to two decades behind developed regions, but worse is falling further behind. We also compare our results with other measures of countries' performance such as International bandwidth, and human development indices. We will then focus on the situation in Africa in terms of challenges, opportunities, connectivity, comparisons with other regions, costs, and possible future developments.

Adoption of information and communication technologies and access to the Internet is expanding in Africa, but because of the rapid growth elsewhere, a Digital Divide between Africa and the rest of the world exists, and the gap is growing. In many sub-Saharan African countries, education and research sector suffer some of the worst deficiencies in access to the Internet, despite progress in development of NRENs National Research and Education (cyber) Networks. By contrast, it is widely acknowledged in policy statements from the African Union, the UN, and others that strength in this very sector provides the key to meeting and sustaining Millennium Development Goals. Developed countries with effective cyber-capabilities proclaim the benefits to rich and poor alike arising from the Information Revolution. This is but a dream for many scientists in African institutions. As the world of science becomes increasingly Internet-dependent, so they become increasingly isolated. eGY-Africa is a bottom-up initiative by African scientists and their collaborators to try to reduce this Digital Divide by a campaign of advocacy for better institutional facilities. Four approaches are being taken. The present status of Internet services, problems, and plans are being mapped via a combination of direct measurement of Internet performance (the PingER Project) and a questionnaire-based survey. Information is being gathered on policy statements and initiatives aimed at reducing the Digital Divide, which can be used for arguing the case for better Internet facilities. Groups of concerned scientists are being formed at the national, regional levels in Africa, building on existing networks as much as possible. Opinion in the international science community is being mobilized. Finally, and perhaps most important of all, eGY-Africa is seeking to engage with the many other programs, initiatives, and bodies that share the goal of reducing the Digital Divide either as a direct policy objective, or indirectly as a means to an end, such as the development of an indigenous capability in science and technology for national development. The expectation is that informed opinion from the scientific community at the institutional, national, and international levels can be used to influence the decision makers and donors who are in a position to deliver better Internet capabilities.

The future of Computing in High Energy Physics (HEP) applications depends on both the Network and Grid infrastructure. South Asian countries such as India and Pakistan are making significant progress by building clusters as well as improving their network infrastructure However to facilitate the use of these resources, they need to manage the issues of network connectivity to be among the leading participants in Computing for HEP experiments. In this paper we classify the connectivity for academic and research institutions of South Asia. The quantitative measurements are carried out using the PingER methodology; an approach that induces minimal ICMP traffic to gather active end-to-end network statistics. The PingER project has been measuring the Internet performance for the last decade. Currently the measurement infrastructure comprises of over 700 hosts in more than 130 countries which collectively represents approximately 99% of the world's Internet-connected population. Thus, we are well positioned to characterize the world's connectivity. Here we present the current state of the National Research and Educational Networks (NRENs) and Grid Infrastructure in the South Asian countries and identify the areas of concern. We also present comparisons between South Asia and other developing as well as developed regions. We show that there is a strong correlation between the Network performance and several Human Development indices.