Eun-Sung Jung

Bandwidth Scheduling and Path Computation Algorithms for Connection-Oriented Networks

There has been an increasing number of network deployments that provide dedicated connections through on-demand and in-advance scheduling in support of high- performance applications. We describe algorithms for scheduling and path computations needed for dedicated bandwidth connections for fixed-slot, highest available bandwidth in a given slot, first available slot, and all-available slots computations. These algorithms for bandwidth scheduling are based on extending the classical breadth-first search, Dijkstra, and Bellman-Ford algorithms. We describe a bandwidth management system for UltraScience Net that incorporates implementations of these algorithms.

An Evaluation of In-Advance Bandwidth Scheduling Algorithms for Connection-Oriented Networks

Several bandwidth management systems have been developed to reserve, in advance, dedicated connections for high-performance applications. We describe the in-advance reservation capabilities of these systems as well as the bandwidth scheduling and path computation algorithms used. An analytical and experimental evaluation of these algorithms also is provided. Our experiments indicate that for the fixed-slot problem, the minimum-hop feasible path algorithm proposed by us in [8] maximizes network utilization for large networks while the dynamic adaptive feasible path algorithm proposed in this paper does this for small networks.

A practical Si nanowire technology with nanowire-on-insulator structure for beyond 10nm logic technologies

This paper reports the design and fabrication of a practical Si nanowire (NW) transistor for beyond 10 nm logic devices application. The dependency of the DC and AC performances of Si NW MOSFETs on NW diameter (DNW) and gate oxide thickness has been investigated. A Si NW device with the scaled DNW of 9 nm and thin equivalent oxide thickness (EOT) of 0.9 nm improved both on-current and electrostatic characteristics. Finally, a Nanowire-On-Insulator (NOI) structure has been proposed to enhance the AC performance of a multiple-stacked NWs structure, which improves DC performance but has the issue of high parasitic capacitance. As a result, the simulated AC performance of a triple-NOI structure was improved by around 20% compared to conventional triple NW structure.

Performance evaluation of routing and wavelength assignment algorithms for optical networks

Several routing and wavelength assignment algorithms have been developed to reserve dedicated connections for high-performance applications on optical networks. In this paper, we present an analytical and experimental evaluation of these algorithms. Our experiments indicate that the minimum-hop feasible path algorithm maximizes network utilization. We also present novel algorithms for deferred wavelength assignment that can be used for reducing the space and time requirements of many wavelength assignment algorithms.

Workflow performance improvement using model-based scheduling over multiple clusters and clouds

In recent years, a variety of computational sites and resources have emerged, and users often have access to multiple resources that are distributed. These sites are heterogeneous in nature and performance of different tasks in a workflow varies from one site to another. Additionally, users typically have a limited resource allocation at each site capped by administrative policies. In such cases, judicious scheduling strategy is required in order to map tasks in the workflow to resources so that the workload is balanced among sites and the overhead is minimized in data transfer. Most existing systems either run the entire workflow in a single site or use naive approaches to distribute the tasks across sites or leave it to the user to optimize the allocation of tasks to distributed resources. This results in a significant loss in productivity. We propose a multi-site workflow scheduling technique that uses performance models to predict the execution time on resources and dynamic probes to identify the achievable network throughput between sites. We evaluate our approach using real world applications using the Swift parallel and distributed execution framework. We use two distinct computational environments-geographically distributed multiple clusters and multiple clouds. We show that our approach improves the resource utilization and reduces execution time when compared to the default schedule.

Improving Multisite Workflow Performance Using Model-Based Scheduling

Workflows play an important role in expressing and executing scientific applications. In recent years, a variety of computational sites and resources have emerged, and users often have access to multiple resources that are geographically distributed. These computational sites are heterogeneous in nature and performance of different tasks in a workflow varies from one site to another. Additionally, users typically have a limited resource allocation at each site. In such cases, judicious scheduling strategy is required in order to map tasks in the workflow to resources so that the workload is balanced among sites and the overhead is minimized in data transfer. Most existing systems either run the entire workflow in a single site or use naive approaches to distribute the tasks across sites or leave it to the user to optimize the allocation of tasks to distributed resources. This results in a significant loss in productivity for a scientist. In this paper, we propose a multi-site workflow scheduling technique that uses performance models to predict the execution time on different resources and dynamic probes to identify the achievable network throughput between sites. We evaluate our approach using real world applications in a distributed environment using the Swift distributed execution framework and show that our approach improves the execution time by up to 60% compared to the default schedule.

Toward optimizing disk-to-disk transfer on 100G networks

The recent emergence of ultra high-speed networks up to 100 Gbps has posed numerous challenges and has led to many investigations on efficient protocols to saturate 100 Gbps links. However, end-to-end data transfers involve many compo-nents, not only protocols, affecting overall transfer performance. These components include a disk I10 subsystem, additional computation associated with data streams, and network adapter capacities. For example, achievable bandwidth by TCP may not be implementable if disk I/O or CPU becomes a bottleneck in end-to-end data transfer. In this paper, we first model all the system components involved in end-to-end data transfer as a graph. We then formulate the problem whose goal is to achieve maximum data transfer throughput using parallel data flows. Our contributions lie in how to optimize data transfers considering all the system components involved rather than in accurately modeling all the system components involved. Our proposed formulations and solutions are evaluated through experiments on the ESnet 100G testbed. The experimental results show that our approach is several times faster than Globus Online - 8x faster for datasets with many 10MB files and 3-4x faster for other datasets of larger size files.

Workflow scheduling in e-Science networks

We solve workflow scheduling problems in e-Science networks, whose goal is minimizing either makespan or network resource consumption by jointly scheduling heterogeneous resources such as compute and network resources. We formulate the workflow scheduling problem incorporating multiple paths as a mixed integer linear programming (MILP) and develop several linear programming relaxation heuristics based on this formulation. Our algorithms allow dynamic multiple paths for data transfer between tasks and more flexible resource allocation that may vary over time. We evaluate our algorithms against a well-known list scheduling algorithm in e-Science networks whose size is relatively small. Our simulation results show that our heuristics are fast and work well when communication-to-computation ratios (CCRs) are small. Also, these results show that use of dynamic multiple paths and malleable resource allocation is useful for data intensive applications.

Advance reservation access control using software-defined networking and tokens

Abstract Advance reservation systems allow users to reserve dedicated bandwidth connection resources from advanced high-speed networks. A common use case for such systems is data transfers in distributed science environments in which a user wants exclusive access to the reservation. However, current advance network reservation methods cannot ensure exclusive access of a network reservation to the specific flow for which the user made the reservation. We present here a novel network architecture that addresses this limitation and ensures that a reservation is used only by the intended flow. We achieve this by leveraging software-defined networking (SDN) and token-based authorization. We use SDN to orchestrate and automate the reservation of networking resources, end-to-end and across multiple administrative domains, and tokens to create a strong binding between the user or application that requested the reservation and the flows provisioned by SDN. We conducted experiments on the ESNet 100G SDN testbed, and demonstrated that our system effectively protects authorized flows from competing traffic in the network.

Profiling Optimization for Big Data Transfer over Dedicated Channels

The transfer of big data is increasingly supported by dedicated channels in high-performance networks, where transport protocols play an important role in maximizing application-level throughput and link utilization. The performance of transport protocols largely depend on their control parameter settings, but it is prohibitively time consuming to conduct an exhaustive search in a large parameter space to find the best set of parameter values. We propose FastProf, a stochastic approximation-based transport profiler, to quickly determine the optimal operational zone of a given data transfer protocol/method over dedicated channels. We implement and test the proposed method using both emulations based on real-life performance measurements and experiments over physical connections with short (2ms) and long (380ms) delays. Both the emulation and experimental results show that FastProf significantly reduces the profiling overhead while achieving a comparable level of end-to-end throughput performance with the exhaustive search-based approach.