Cho-Li Wang

Cho-Li Wang

Auto-generated profile, based on publicly available information

Cho-Li Wang

Department of Computer Science, The University of Hong Kong, Hong Kong, China

Research Focus & Metrics

Academic Career Length

33Years (1994-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 May 2012 in IEEE Transactions on Parallel and Distributed Systems
Verified
Journal Article
Closed access
SDSheng Di
Sheng Di
CWCho-Li Wang
Cho-Li Wang
By leveraging virtual machine (VM) technology which provides performance and fault isolation, cloud resources can be provisioned on demand in a fine grained, multiplexed manner rather than in monolithic pieces. By integrating volunteer computing into cloud architectures, we envision a gigantic self-organizing cloud (SOC) being formed to reap the huge potential of untapped commodity computing power over the Internet. Toward this new architecture where each participant may autonomously act as both resource consumer and provider, we propose a fully distributed, VM-multiplexing resource allocation scheme to manage decentralized resources. Our approach not only achieves maximized resource utilization using the proportional share model (PSM), but also delivers provably and adaptively optimal execution efficiency. We also design a novel multiattribute range query protocol for locating qualified nodes. Contrary to existing solutions which often generate bulky messages per request, our protocol produces only one lightweight query message per task on the Content Addressable Network (CAN). It works effectively to find for each task its qualified resources under a randomized policy that mitigates the contention among requesters. We show the SOC with our optimized algorithms can make an improvement by 15-60 percent in system throughput than a P2P Grid model. Our solution also exhibits fairly high adaptability in a dynamic node-churning environment.

17 November 2013
Verified
Conference Paper
Open access
SDSheng Di
Sheng Di
YRYves Robert
Yves Robert
FVFrédéric Vivien
Frédéric Vivien
DKDerrick Kondo
Derrick Kondo
CWCho-Li Wang
Cho-Li Wang
FCFranck Cappello
Franck Cappello
In this paper, we aim at optimizing fault-tolerance techniques based on a checkpointing/restart mechanism, in the context of cloud computing. Our contribution is three-fold. (1) We derive a fresh formula to compute the optimal number of checkpoints for cloud jobs with varied distributions of failure events. Our analysis is not only generic with no assumption on failure probability distribution, but also attractively simple to apply in practice. (2) We design an adaptive algorithm to optimize the impact of checkpointing regarding various costs like checkpointing/restart overhead. (3) We evaluate our optimized solution in a real cluster environment with hundreds of virtual machines and Berkeley Lab Checkpoint/Restart tool. Task failure events are emulated via a production trace produced on a large-scale Google data center. Experiments confirm that our solution is fairly suitable for Google systems. Our optimized formula outperforms Young's formula by 3-10 percent, reducing wall-clock lengths by 50-100 seconds per job on average.

1 October 2000 in Journal of Parallel and Distributed Computing
Verified
Journal Article
Closed access
MMMatchy J.M. Ma
Matchy J.M. Ma
CWCho-Li Wang
Cho-Li Wang
FLFrancis C.M. Lau
Francis C.M. Lau

1 July 1996 in Proceedings of the IEEE
Verified
Journal Article
Open access
CWCho-Li Wang
Cho-Li Wang
PBP.B. Bhat
P.B. Bhat
VPV.K. Prasanna
V.K. Prasanna
Vision is a challenging application for high-performance computing (HPC). Many vision tasks have stringent latency and throughput requirements. Further, the vision process has a heterogeneous computational profile. Low-level vision consists of structured computations, with regular data dependencies. The subsequent, higher level operations consist of symbolic computations with irregular data dependencies. Over the years, many approaches to high-speed vision have been pursued. VLSI hardware solutions such as ASIC's and digital signal processors (DSP's) have provided good processing speeds on structured low-level vision tasks. Special purpose systems for vision have also been designed. Currently, there is growing interest in using general purpose parallel systems for vision problems. These systems offer advantages of higher performance, sofavare programmability, generality, and architectural flexibility over the earlier approaches. The choice of low-cost commercial-off-theshelf (COTS) components as building blocks for these systems leads to easy upgradability and increased system life. The main focus of the paper is on effectively using the COTSbased general purpose parallel computing platforms to realize high-speed implementations of vision tasks. Due to the successful use of the COTS-based systems in a variety of high performance applications, it is attractive to consider their use for vision applications as well. However, the irregular data dependencies in vision tasks lead to large communication overheads in the HPC systems. At the University of Southern California, our research efforts have been directed toward designing scalable parallel algorithms for vision tasks on the HPC systems. In our approach, we use the message passing programming model to develop portable code. Our algorithms are specified using C and MPI. In this paper, we summarize our efforts, and illustrate our approach using several example vision tasks. To facilitate the analysis and development of scalable algorithms, a realistic computational model of the parallel system must be used. Several such models have been proposed in the literature. We use the General-purpose Distributed Memory (GDM) model which is a simple but realistic model of state-of-theart parallel machines. Using the GDM model, generic algorithmic techniques such as data remapping, overlapping of communication with computation, message packing, asynchronous execution, and communication scheduling are developed. Using these techniques, we have developed scalable algorithms for many vision tasks. For instance, a scalable algorithm for linear approximation has been developed using the asynchronous execution technique. Using this algorithm, linear feature extraction can be performed in 0.065 s on a 64 node SP-2 for a 512 × 512 image. A serial implementation takes 3.45 s for the same task. Similarly, the communication scheduling and decomposition techniques lead to a scalable algorithm for the line grouping task. We believe that such an algorithmic approach can result in the development of scalable and portable solutions for vision tasks. © 1996 IEEE Publisher Item Identifier S 0018-9219(96)04992-4.published_or_final_versio

22 October 2012 in IEEE Transactions on Parallel and Distributed Systems
Verified
Journal Article
Closed access
SDSheng Di
Sheng Di
CWCho-Li Wang
Cho-Li Wang
With virtual machine (VM) technology being increasingly mature, compute resources in cloud systems can be partitioned in fine granularity and allocated on demand. We make three contributions in this paper: 1) We formulate a deadline-driven resource allocation problem based on the cloud environment facilitated with VM resource isolation technology, and also propose a novel solution with polynomial time, which could minimize users' payment in terms of their expected deadlines. 2) By analyzing the upper bound of task execution length based on the possibly inaccurate workload prediction, we further propose an error-tolerant method to guarantee task's completion within its deadline. 3) We validate its effectiveness over a real VM-facilitated cluster environment under different levels of competition. In our experiment, by tuning algorithmic input deadline based on our derived bound, task execution length can always be limited within its deadline in the sufficient-supply situation; the mean execution length still keeps 70 percent as high as user-specified deadline under the severe competition. Under the original-deadline-based solution, about 52.5 percent of tasks are completed within 0.95-1.0 as high as their deadlines, which still conforms to the deadline-guaranteed requirement. Only 20 percent of tasks violate deadlines, yet most (17.5 percent) are still finished within 1.05 times of deadlines.

1 December 2011
Verified
Conference Paper
Closed access
RMRicky K. K. Ma
Ricky K. K. Ma
KLKing Tin Lam
King Tin Lam
CWCho-Li Wang
Cho-Li Wang
Cloud computing augments applications with ease-of-access to the enormous resources on the Internet. Combined with mobile computing technologies, mobile applications can exploit the Cloud everywhere by statically distributing code segments or dynamically migrating running processes onto cloud services. Existing migration techniques are however too coarse-grained for mobile devices, so the overheads often offset the benefits of migration. To build a truly elastic mobile cloud computing infrastructure, we introduce eXCloud (eXtensible Cloud) - a middleware system with multi-level mobility support, ranging from as coarse as a VM instance to as fine as a runtime stack frame, and allows resources to be integrated and used dynamically. In eXCloud, a stack-on-demand (SOD) approach is used to support computation mobility throughout the mobile cloud environment. The approach is fully adaptive, goal-driven and transparent. By downward task migration, applications running on the cloud nodes can exploit or take control of special resources in mobile devices such as GPS and cameras. With a restorable MPI layer, task migrations of MPI parallel programs can happen between cloud nodes or be initiated from a mobile device. Our evaluation shows that SOD outperforms several existing migration mechanisms in terms of migration overhead and latency. All our techniques result in better resource utilization through task migrations among cloud nodes and mobile nodes.