Verdi March

Bio: Verdi March is an academic researcher from Hewlett-Packard. The author has contributed to research in topics: Cloud computing & Overlay network. The author has an hindex of 12, co-authored 25 publications receiving 499 citations. Previous affiliations of Verdi March include Sun Microsystems & National University of Singapore.

Evaluating and Improving the Performance and Scheduling of HPC Applications in Cloud

Abstract: Cloud computing is emerging as a promising alternative to supercomputers for some high-performance computing (HPC) applications. With cloud as an additional deployment option, HPC users and providers are faced with the challenges of dealing with highly heterogeneous resources, where the variability spans across a wide range of processor configurations, interconnects, virtualization environments, and pricing models. In this paper, we take a holistic viewpoint to answer the question— why and who should choose cloud for HPC, for what applications, and how should cloud be used for HPC? To this end, we perform comprehensive performance and cost evaluation and analysis of running a set of HPC applications on a range of platforms, varying from supercomputers to clouds. Further, we improve performance of HPC applications in cloud by optimizing HPC applications’ characteristics for cloud and cloud virtualization mechanisms for HPC. Finally, we present novel heuristics for online application-aware job scheduling in multi-platform environments. Experimental results and simulations using CloudSim show that current clouds cannot substitute supercomputers but can effectively complement them. Significant improvement in average turnaround time (up to 2X)and throughput (up to 6X) can be attained using our intelligent application-aware dynamic scheduling heuristics compared tosingle-platform or application-agnostic scheduling.

The Who, What, Why, and How of High Performance Computing in the Cloud

Abstract: Cloud computing is emerging as an alternative to supercomputers for some of the high-performance computing (HPC) applications that do not require a fully dedicated machine. With cloud as an additional deployment option, HPC users are faced with the challenges of dealing with highly heterogeneous resources, where the variability spans across a wide range of processor configurations, interconnections, virtualization environments, and pricing rates and models. In this paper, we take a holistic viewpoint to answer the question - why and who should choose cloud for HPC, for what applications, and how should cloud be used for HPC? To this end, we perform a comprehensive performance evaluation and analysis of a set of benchmarks and complex HPC applications on a range of platforms, varying from supercomputers to clouds. Further, we demonstrate HPC performance improvements in cloud using alternative lightweight virtualization mechanisms - thin VMs and OS-level containers, and hyper visor- and application-level CPU affinity. Next, we analyze the economic aspects and business models for HPC in clouds. We believe that is an important area that has not been sufficiently addressed by past research. Overall results indicate that current public clouds are cost-effective only at small scale for the chosen HPC applications, when considered in isolation, but can complement supercomputers using business models such as cloud burst and application-aware mapping.

Survey on Parallel Programming Model

Abstract: The development of microprocessors design has been shifting to multi-core architectures. Therefore, it is expected that parallelism will play a significant role in future generations of applications. Throughout the years, there has been a myriad number of parallel programming models proposed. In choosing a parallel programming model, not only the performance aspect is important, but also qualitative the aspect of how well parallelism is abstracted to developers. A model with a well abstraction of parallelism leads to a higher application-development productivity. In this paper, we propose seven criteria to qualitatively evaluate parallel programming models. Our focus is on how parallelism is abstracted and presented to application developers. As a case study, we use these criteria to investigate six well-known parallel programming models in the HPC community.

Exploring the performance and mapping of HPC applications to platforms in the cloud

Abstract: This paper presents a scheme to optimize the mapping of HPC applications to a set of hybrid dedicated and cloud resources. First, we characterize application performance on dedicated clusters and cloud to obtain application signatures. Then, we propose an algorithm to match these signatures to resources such that performance is maximized and cost is minimized. Finally, we show simulation results revealing that in a concrete scenario our proposed scheme reduces the cost by 60% at only 10-15% performance penalty vs. a non optimized configuration. We also find that the execution overhead in cloud can be minimized to a negligible level using thin hypervisors or OS-level containers.

Designing an Super-Peer Network

Abstract: A super-peer is a node in a peer-to-peer network that operates both as a server to a set of clients, and as an equal in a network of super-peers. Super-peer networks strike a balance between the efficiency of centralized search, and the autonomy, load balancing and robustness to attacks provided by distributed search. Furthermore, they take advantage of the heterogeneity of capabilities (e.g., bandwidth, processing power) across peers, which recent studies have shown to be enormous. Hence, new and old P2P systems like KaZaA and Gnutella are adopting super-peers in their design. Despite their growing popularity, the behavior of super-peer networks is not well understood. For example, what are the potential drawbacks of super-peer networks? How can super-peers be made more reliable? How many clients should a super-peer take on to maximize efficiency? we examine super-peer networks in detail, gaming an understanding of their fundamental characteristics and performance tradeoffs. We also present practical guidelines and a general procedure for the design of an efficient super-peer network.

A Survey of Mobile Cloud Computing Application Models

Abstract: Smart phones are now capable of supporting a wide range of applications, many of which demand an ever increasing computational power. This poses a challenge because smart phones are resource-constrained devices with limited computation power, memory, storage, and energy. Fortunately, the cloud computing technology offers virtually unlimited dynamic resources for computation, storage, and service provision. Therefore, researchers envision extending cloud computing services to mobile devices to overcome the smartphones constraints. The challenge in doing so is that the traditional smartphone application models do not support the development of applications that can incorporate cloud computing features and requires specialized mobile cloud application models. This article presents mobile cloud architecture, offloading decision affecting entities, application models classification, the latest mobile cloud application models, their critical analysis and future research directions.

Heterogeneity in Mobile Cloud Computing: Taxonomy and Open Challenges

Abstract: The unabated flurry of research activities to augment various mobile devices by leveraging heterogeneous cloud resources has created a new research domain called Mobile Cloud Computing (MCC). In the core of such a non-uniform environment, facilitating interoperability, portability, and integration among heterogeneous platforms is nontrivial. Building such facilitators in MCC requires investigations to understand heterogeneity and its challenges over the roots. Although there are many research studies in mobile computing and cloud computing, convergence of these two areas grants further academic efforts towards flourishing MCC. In this paper, we define MCC, explain its major challenges, discuss heterogeneity in convergent computing (i.e. mobile computing and cloud computing) and networking (wired and wireless networks), and divide it into two dimensions, namely vertical and horizontal. Heterogeneity roots are analyzed and taxonomized as hardware, platform, feature, API, and network. Multidimensional heterogeneity in MCC results in application and code fragmentation problems that impede development of cross-platform mobile applications which is mathematically described. The impacts of heterogeneity in MCC are investigated, related opportunities and challenges are identified, and predominant heterogeneity handling approaches like virtualization, middleware, and service oriented architecture (SOA) are discussed. We outline open issues that help in identifying new research directions in MCC.

Cloud-Based Augmentation for Mobile Devices: Motivation, Taxonomies, and Open Challenges

Abstract: Recently, Cloud-based Mobile Augmentation (CMA) approaches have gained remarkable ground from academia and industry. CMA is the state-of-the-art mobile augmentation model that employs resource-rich clouds to increase, enhance, and optimize computing capabilities of mobile devices aiming at execution of resource-intensive mobile applications. Augmented mobile devices envision to perform extensive computations and to store big data beyond their intrinsic capabilities with least footprint and vulnerability. Researchers utilize varied cloud-based computing resources (e.g., distant clouds and nearby mobile nodes) to meet various computing requirements of mobile users. However, employing cloud-based computing resources is not a straightforward panacea. Comprehending critical factors (e.g., current state of mobile client and remote resources) that impact on augmentation process and optimum selection of cloud-based resource types are some challenges that hinder CMA adaptability. This paper comprehensively surveys the mobile augmentation domain and presents taxonomy of CMA approaches. The objectives of this study is to highlight the effects of remote resources on the quality and reliability of augmentation processes and discuss the challenges and opportunities of employing varied cloud-based resources in augmenting mobile devices. We present augmentation definition, motivation, and taxonomy of augmentation types, including traditional and cloud-based. We critically analyze the state-of-the-art CMA approaches and classify them into four groups of distant fixed, proximate fixed, proximate mobile, and hybrid to present a taxonomy. Vital decision making and performance limitation factors that influence on the adoption of CMA approaches are introduced and an exemplary decision making flowchart for future CMA approaches are presented. Impacts of CMA approaches on mobile computing is discussed and open challenges are presented as the future research directions.