Haiyang Qian

A hierarchical model to evaluate quality of experience of online services hosted by cloud computing

As online service providers utilize cloud computing to host their services, they are challenged by evaluating the quality of experience and designing redirection strategies in this complicated environment. We propose a hierarchical modeling approach that can easily combine all components of this environment. Identifying interactions among the components is the key to construct such models. In this particular environment, we first construct four sub-models: an outbound bandwidth model, a cloud computing availability model, a latency model and a cloud computing response time model. Then we use a redirection strategy graph to glue them together. We also introduce an all-in-one barometer to ease the evaluation. The numeric results show that our model serves as a very useful analytical tool for online service providers to evaluate cloud computing providers and design redirection strategies.

The great plains environment for network innovation (GpENI):a programmable testbed for future internet architecture research

The Great Plains Environment for Network Innovation – GpENI is an international programmable network testbed centered on a regional optical network in the Midwest US, providing flexible infrastructure across the entire protocol stack. The goal of GpENI is to build a collaborative research infrastructure enabling the community to conduct experiments in future Internet architecture. GpENI is funded in part by the US National Science Foundation GENI (Global Environments for Network Innovation) program and by the EU FIRE (Future Internet Research and Experimentation) Programme, and is affiliated with a project funded by the NSF FIND (Future Internet Design) Program.

On energy-aware aggregation of dynamic temporal demand in cloud computing

The proliferation of cloud computing faces social and economic concerns on energy consumption. We present formulations for cloud servers to minimize energy consumption as well as server hardware cost under three different models (homogeneous, heterogeneous, mixed hetero-homogeneous clusters) by considering dynamic temporal demand. To be able to compute optimal configurations for large scale clouds, we then propose static and dynamic aggregation methods, which come at the additional cost on energy consumption; however, they still result in significant savings compared to the scenario when all servers are on during the entire duration. Our studies show that the homogeneous model takes four time less computational time than the heterogeneous model. The dynamic aggregation scheme results in 8% to 40% savings over the static aggregation scheme when the degree of aggregation is high.

Service management architecture and system capacity design for PhoneFactor™—A two-factor authentication service

PhoneFactor™ is a token-less two-factor authentication service for user remote logons [13]. This allows users of an organization to be authenticated through an automated phone call to the users phone before access is allowed. In this paper, we present the service management architecture of PhoneFactor that depends on both the Internet and the public switched telephone network (PSTN), and we identify two key quality of service parameters, the system response time and call blocking probability, where the latter can impact the former. Furthermore, through traffic analysis of the measurement data from the deployed PhoneFactor service, we found that the inter-arrival time of requests follows the Generalized Pareto distribution while the system response time and the call duration (for the authentication part through the phone call) follow the log-normal distribution. Given these distributions, we then present system capacity design methodologies by comparing them to known results for systems that are analytically derivable.

Optimal Resource Provisioning and the Impact of Energy-Aware Load Aggregation for Dynamic Temporal Workloads in Data Centers

An important goal of data center providers is to minimize their operational cost, which reflected through the wear-and-tear cost and the energy consumption cost. In this paper, we present optimization formulations to minimize the cost of ownership in terms of server energy consumption and serverwear-and-tear cost under three different data center server setups (homogeneous, heterogeneous, and hybrid hetero-homogeneous clusters) for dynamic temporal workloads. Our studies show that the homogeneous model takes significantly less computational time than the heterogeneous model (by an order of magnitude). To compute optimal configurations in near real time for large-scale data centers, we propose two modes for using our models: aggregation by maximum (preserves workload deadline) and aggregation by mean (relaxes workload deadline). In addition, we propose two aggregation methods for use in each of the two modes: static (periodic) aggregation and dynamic (aperiodic)aggregation. We found that in the aggregation by maximum mode, dynamic aggregation resulted in cost savings of up to approximately 18% over the static aggregation. In the aggregation by mean mode, dynamic aggregation saved up to approximately a 50% workload rearrangement compared with the static aggregationby mean mode.

DReaM-Cache: Distributed Real-Time Transaction Memory Cache to support two-factor authentication services and its reliability

PhoneFactor is a two-factor authentication service that combines the knowledge-based authenticator with an object-based authenticator in which the object-based authenticator has advantages of hardware token-based systems without requiring the administration overhead of distribution of special hardware to every user in advance. For transaction scalability of this service, we propose a Distributed Real-Time Transaction Memory Cache (DReaM-Cache) approach. In this paper, we present the basic architecture and present reliability models for different data centers and node reliability factors, and show how reliability can be accomplished.