Large-scale data transport for data-intensive sciences is a complex multidimensional challenge. The challenge includes optimizing the end-to-end Big Data movement performance in real-time, supporting direct remote data access using NVMe over Fabrics (NVMeoF) and deploying to existing research platforms. AIDTN is the first effort to provide a unique AI system designed to incorporate NVMe over Fabrics (NVMeoF) and optimize coordination among multiple components supporting large-scale, multi-domain Wide Area Network (WAN) data-intensive science. AIDTN's research objective is to integrate next-generation storage architecture using NVMeoF, specialized network design using high-performance network appliances, Data Transfer Nodes (DTNs), catalysts in driving data transport, and a unique AI system explicitly designed for high-performance data movement challenges. AIDTN is the first system that uses network and system features to predict the end-to-end performance of high-performance data movement and further extends the model with NVMe-specific features for NVMeoF remote data access. As a result, AIDTN improves data movement performance by up to 284% while minimizing packet loss compared to other heuristics approaches. It also has a prediction error rate as low as 0.16 compared to AI models with the only network (error rate = 0.29) or network and system features (error rate = 0.19).

The goal of P4MT is to leverage the role-based control mechanism in P4Runtime Specification to further enable multi-tenancy on a single P4 switch. The evaluation results show that P4MT consumes only a small percentage of ASIC space and causes a negligible increment in data plane/control plane latency. International P4 Experimental Networks (i-P4EN) has been created to realize the potential of P4 experimental networks. To perform multiple experiments for research groups, multitenancy is required to increase the efficiency of the testbed. By modifying the P4Runtime design and P4 pipeline, we designed and implemented multi-tenancy on a P4 switch. With P4MT, each tenant can select its own P4 pipeline and control the packet processing without intervening with the packet processing of other tenants. Future work includes supporting P4 Externs for programmable data plane networking, multi-tenant networking and automatic pipeline migration on P4MT.

To realize P4 language benefits: “Protocol Independent, Target Independent, Field Reconfigurable” [1] in a network research and development environment, a few P4 research institutions around the world formed a partnership to design and implement the international P4 Experimental Networks (Figure 1). The participating institutions can share distributed P4 resources over international research and education networks, as well as initiate international P4 research collaboration projects. Initial network scenarios in this research testbed are Software-Defined Network Exchanges (SDXs), Network and Cloud Testbed Networks, Science Networks, and Campus Research Networks. These four network scenarios require multi-tenant support. Current P4Runtime specifications include a Multi-Controller design for twofold: Control Plane Partition and Redundancy [2]. However, the current P4Runtime does not implement Control Plane partition to support multi-tenant service on a switch. This project modified the current design to support a provider and multiple tenant model for the four international P4 Experimental Networks (i-P4EN)scenarios and other applications.

Computer Science experimental testbeds allow investigators to explore a broad range of different state-of-the-art hardware options, assess scalability of their systems, and provide conditions that allow deep reconfigurability and isolation so that one user does not impact the experiments of another. An experimental testbed is also in a unique position to support methods facilitating experiment analysis and improve repeatability and reproducibility of experiments. Providing these capabilities at least partially within a commodity framework improves the sustainability of systems experiments and thus makes them available to a broader range of experimenters.

This paper discusses the evolution of optical networks from foundational data transport utilities and infrastructure to true enablers of next-generation services and facilities, specifically diverse rich digital media services, which are being produced, processed, and consumed by an increasingly sophisticated variety of end users. This evolution has been enabled by several major recent advances in optical networking, including the deployment of high-speed optical connections from core networks to enterprise and residential users; the dramatic increase in optical signal speed and reach; and the improvement in control and management planes to support increasingly diverse types of traffic and services. We describe two distinctive areas of broad impact of optical network technologies related to rich media services: high-resolution scientific visualization and high-quality, real-time consumer-driven media production and distribution. Finally, we discuss key trends that may develop next as optical networks further evolve shaped by both rich digital media service requirement and technology innovation.

For many years, experimental network research test beds have been designed and implemented to explore theoretical concepts at scale, progressing beyond ideas that initially may have been explored through lab simulation and modeling. However, traditionally, such test beds almost always were created for specific, highly defined projects, and they were designed to last for fairly short periods of time. More recently, network research communities have been designing, implementing, and operating new types of large scale experimental network test beds, including those global in scope, which are intended to be long term persistent resources and can support multiple types of experiments. These test beds reflect recent innovations in architecture and in core technologies and also the declining cost curves of basic components, a trend which is reducing the economic barriers to developing test beds. This paper presents several key ideas motivating the creation of next generation test beds and it describes design concepts, core technologies, and operational issues. This paper also describes a number of recent and current large scale test beds that have been created to advance the state of communications, especially those designed to assist in providing a transition to the future Internet.

The Open Science Data Cloud is a distributed cloud based infrastructure for managing, analyzing, archiving and sharing scientific datasets. We introduce the Open Science Data Cloud, give an overview of its architecture, provide an update on its current status, and briefly describe some research areas of relevance.

National Center for Data Mining at UICIn our SC06 BWC entry, we will transfer SDSS (Sloan Digital Sky Survey) Data Release 5 (DR5) between the SC06 show floor in Tampa and one of the NCDM labs on the UIC campus. We will use SECTOR, our newly developed distributed data space management system, to transfer DR5 in parallel between two Linux clusters in Tampa and Chicago, respectively. SECTOR transparently manages the file locating and data moving, while it employs UDT for actual data transfer. The data transfer will be from disk to disk over a 10Gb/s shared, router link between SC06 and UIC, via StarLight. We expect to reach 5Gb/s disk-to-disk data transfer rate between the two sites.

Advances in Grid technology enable the deployment of data-intensive distributed applications, which require moving terabytes or even petabytes of data between data banks. The current underlying networks cannot provide dedicated links with adequate end-to-end sustained bandwidth to support the requirements of these Grid applications. DWDM-RAM is a novel service-oriented architecture, which harnesses the enormous bandwidth potential of optical networks and demonstrates their on-demand usage on the OMNInet. Preliminary experiments suggest that dynamic optical networks, such as the OMNInet, are the ideal option for transferring such massive amounts of data. DWDM-RAM incorporates an OGSI/OGSA compliant service interface and promotes greater convergence between dynamic optical networks and data intensive Grid computing.

Next generation applications and architectures (for example, Grids) are driving radical changes in the nature of traffic, service models, technology, and cost, creating opportunities for an advanced communications infrastructure to tackle next generation data services. To take advantage of these trends and opportunities, research communities are creating new architectures, such as the Open Grid Service Architecture (OGSA), which are being implemented in new prototype advanced infrastructures. The DWDM-RAM project, funded by DARPA, is actively addressing the challenges of next generation applications. DWDM-RAM is an architecture for data-intensive services enabled by next generation dynamic optical networks. It develops and demonstrates a novel architecture for new data communication services, within the OGSA context, that allows for managing extremely large sets of distributed data. Novel features move network services beyond notions of the network as a managed resource, for example, by including capabilities for dynamic on-demand provisioning and advance scheduling. DWDM-RAM encapsulates optical network resources (Lambdas, lightpaths) into a Grid service and integrates their management within the Open Grid Service Architecture. Migration to emerging standards such as WS-Resource Framework (WS-RF) should be straightforward. In initial applications, DWDM-RAM targets specific data-intensive services such as rapid, massive data transfers used by large scale eScience applications, including: high-energy physics, geophysics, life science, bioinformatics, genomics, medical morphometry, tomography, microscopy imaging, astronomical and astrophysical imaging, complex modeling, and visualization.