Cheng Luo

The University of Tokyo, Edo, Tōkyō, Japan

Are you Cheng Luo?

Claim your profile

Publications (9)1.14 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: In virtualization technology field, researches mainly focus on strengthening the isolation barrier between virtual machines (VMs) that are co-resident within a single physical machine. At the same time, there are many kinds of distributed communication-intensive applications such as web services, transaction processing, graphics rendering and high performance grid applications, which need to communicate with other virtual machines at the same platform. Unfortunately, current inter-VM communication method cannot adequately satisfy the requirement of such applications. In this paper, we present the design and implementation of a high performance inter-VM communication method called IVCOM based on Xen virtual machine environment. In para-virtualization, IVCOM achieves high performance by bypassing some protocol stacks and privileged domain, shunning page flipping and providing a direct and high-performance communication path between VMs residing in the same physical machine. But in full-virtualization, IVCOM applies a direct communication channel between domain 0 and Hardware Virtualization based VM (HV2M) and can greatly reduce the VM entry/exit operations, which has improved the HV2M performance. In the evaluation of para-virtualization consisting of a few of benchmarks, we observe that IVCOM can reduce the inter-VM round trip latency by 70% and increase throughput by up to 3 times, which prove the efficiency of IVCOM in para-virtualized environment. In the full-virtualized one, IVCOM can reduce 90% VMX transition operations in the communication between domain 0 and HV2M.
    Journal of Systems and Software 02/2013; 86(2):367–376. · 1.14 Impact Factor
  • Liang Zhang, Yuebin Bai, Cheng Luo
    [Show abstract] [Hide abstract]
    ABSTRACT: In virtualization environment, all physical hardware resource is maintained by virtual machine monitor (VMM). And as more and more applications and software deployed on virtual machine are communication intensive, they have a demand for communicate with each other. In-built communication ability between virtual machines is a necessary component of a mature virtual environment and is often easy to use, however, the performance of native in-built inter virtual machine communication is often not so good. At the same time, in virtual environment such as Xen hypervisor, a delicate approach, shared memory and event channel mechanism is provided. And this mechanism can be utilized for inter domain communication but not so easy to use. And to get a balance between performance and convenience, in this paper, we design and implement idsocket, an API suite for inter domain communication based on Xen using shared memory and event channel mechanism, bypassing the traditional front-back driver model. Benchmark evaluations and program tests have demonstrated that it has a better performance and a lower resource cost than the in-built front-back driver model for inter domain communication in Xen. Meanwhile, the rule and philosophy behind the design and implementation of idsocket is almost universal in any type of virtual machine.
    Algorithms and Architectures for Parallel Processing, 10th International Conference, ICA3PP 2010, Busan, Korea, May 21-23, 2010. Proceedings. Part I; 01/2010
  • [Show abstract] [Hide abstract]
    ABSTRACT: In virtualization technology domain, researches mainly focus on strengthening the isolation barrier between virtual machines (VMs) that are co-resident within a single physical machine. At the same time, there are many kinds of communication intensive distributed applications such as web services, transaction processing, graphics rendering and high performance grid applications, which need to communicate with each other on the co-resident VMs. Current inter-VM communication mechanisms can’t adequately satisfy the requirement of such applications. In this paper, we present the design and implementation of a high performance inter-VM communication mechanism called IVCOM in Xen virtual machine environment. We propose IVCOM in para-virtualization and also extend for full-virtualization. As a result of our survey, in Para-virtualization, there are mainly three kinds of overheads that contribute to the poor performance: the TCP/IP processing cost in each domain, page flipping overhead and long communication path between both sides of the socket. IVCOM achieves high performance by bypassing protocol stacks, shunning page flipping and providing a direct and high performance communication path between VMs residing with the same physical machine. And in Full-virtualization, frequent mode tuning between root mode and non-root mode import too much overhead. IVCOM applies a direct communication channel between domain 0 and hardware virtual VM (HVM) and can greatly reduce the VM entry/exit operations which can improve the HVM performance. In our evaluation, we observe that IVOCM can reduce the inter-VM round trip latency by up to 3 times and increase throughput by up to 3 times which prove the efficiency of IVCOM in para-virtualized environment. In full-virtualized environment, IVCOM can greatly reduce mode tuning times in the communication between domain 0 and HVM.
    Algorithms and Architectures for Parallel Processing, 10th International Conference, ICA3PP 2010, Busan, Korea, May 21-23, 2010. Proceedings. Part I; 01/2010
  • [Show abstract] [Hide abstract]
    ABSTRACT: The master/worker pattern is widely used to construct the cross-domain, large scale computing infrastructure. The applications supported by this kind of infrastructure usually features long-running, speculative execution etc. Fault recovery mechanism is significant to them especially in the wide area network environment, which consists of error prone components. Inter-node cooperation is urgent to make the recovery process more efficient. The traditional log-based rollback recovery mechanism which features independent recovery cannot fulfill the global cooperation requirement due to the waste of bandwidth and slow application data transfer which is caused by the exchange of a large amount of logs. In this paper, we propose a two-phase log-based recovery mechanism which is of merits such as space saving and global optimization and can be used as a complement of the current log-based rollback recovery approach in some specific situations. We have demonstrated the use of this mechanism in the Drug Discovery Grid environment, which is supported by China National Grid. Experiment results have proved efficiency of this mechanism.
    Parallel and Distributed Processing with Applications, 2009 IEEE International Symposium on; 09/2009
  • [Show abstract] [Hide abstract]
    ABSTRACT: Grid computing has gained wide interests for being an innovative approach that leverages existing IT infrastructure for large volume computing/data processing. However, a challenging issue faced by Grid computing is that the available resources often do not match the needs of a specific application requirement because of their heterogeneity, distributive and dynamic nature. The mismatching between resources and application requirement leads to resources underutilization and user frustration. Cloud computing in which dynamically scalable and often virtualized resources are provided as services can deal with the problem well. In this paper we propose a mechanism of Reliable Elastic Computing Service (R-ECS) based on the Cloud computing concept. ECS aggregates resources to construct a virtual computing environment based on virtualization technologies which reduce the coupling between software and hardware. In addition, it provides services supporting on-demand user-customized resources management, which achieves better resource utilization and user satisfaction. Reliability issue is also addressed by appropriate exception detection and fault recovery. Performance evaluation results show that the user-customized virtual computing environment is efficient and increases resources utilization.
    Proceedings of the 2nd International Conference on Interaction Sciences: Information Technology, Culture and Human 2009, Seoul, Korea, 24-26 November 2009; 01/2009
  • Cong Xu, Yuebin Bai, Cheng Luo
    [Show abstract] [Hide abstract]
    ABSTRACT: As multi-core processors become increasingly mainstream, architects have likewise become more interested in how best to make use of the computing capacity of the CPU, for instance, through multiple simultaneous threads or processes of execution with OpenMP or MPI. At the same time, the increasingly mature and prevailing virtualization technique in server consolidation and HPC promotes the emergence of a large number of virtual SMP servers. Therefore, whether the parallel program can run in the virtual machine environment efficiently or not is a topic of concern. In this paper, we investigate the performance of three typical parallel programming paradigms, including OpenMP, MPI, and Hybrid of OpenMP and MPI in the popular, open-source, Xen virtualization system. The results show that the performance of the traditional parallel program in Xen VMs is close to it in native, non-virtualized environment, if there is little communication or synchronization between threads or processes. In most cases, without excessive IO access, we can get an ideal speedup in a SMP VM or virtual cluster, which is close to linearity when the total virtual CPUs (vCPUs) number is not larger than the number of Physical CPUs (pCPUs). And the pure MPI implementation shows the best scalability and stability in virtual machine environment compared with the other two paradigms.
    NPC 2009, Sixth IFIP International Conference on Network and Parallel Computing, Gold Coast, Australia, October 19-21, 2009; 01/2009
  • [Show abstract] [Hide abstract]
    ABSTRACT: With the resurgence of virtualization technologies and the development of multi-core technologies, the combination of the two becomes a trend. Therefore, inter-VM communication becomes a key part in how to improve the performance of virtual machines (VMs) basing on multi-core platform. In this paper, we first analyze the characteristics of multi-core tasks and the properties of virtual machine environment, and then classify processor cores into two categories basing on their different functions. According to the classification, we design an inter-VM communication mechanism with multi-core platform. It discards the traditional communication path between VMs which needs to via a trusted VM, sets up communication channels between virtual CPUs in different VMs and uses shared memory space to implement high-throughput communication of inter-VM. Experiment results have proved the efficiency of them.
    IEEE 15th International Conference on Parallel and Distributed Systems, ICPADS 2009, 8-11 December 2009, Shenzhen, China; 01/2009
  • [Show abstract] [Hide abstract]
    ABSTRACT: With the increasing number of cores integrated on a single chip, inter-core communication becomes more important in how to effectively support applications on multi-core platform. In this paper, we analyse the characteristics of multi-core tasks and classify processor cores into two categories based on their different functions: control core and computing core. According to the classification, we design a novel inter-core communication mechanism which includes three communication paths. With these paths, we can improve the utilization of multi-core by migrating the I/O part of applications from computing core to control core. Experiment results have proved the efficiency of them.
    Information Science and Engineering, International Conference on. 01/2009;
  • Cong Xu, Yuebin Bai, Cheng Luo, Xiang Ni
    [Show abstract] [Hide abstract]
    ABSTRACT: As multi-core processors become increasingly the mainstream, people have likewise become more interested in how best to make use of the computing capacity of the CPU. Although many methods, running multi-thread application for example, have been adopted to increase the CPU utilization, most multi-core PC's and workstation's CPU cycles are idle, even during peak hours. So it is an efficient solution to help a personal user to build his own small non-dedicated cluster by collecting some idle PC in a LAN. In order to improve the utilization of the multi-core processor and shield the heterogeneity of different platform, virtual machine (VM) technology can be applied to partition the resource of each computer, changing a physical node into several homogeneous virtual nodes. This personal virtual cluster (PVC) can be created, managed, and released by a personal user, and run some computationally intensive parallel program such as application with MPI during some temporary time. In this paper we present a prototype of PVC with the popular, open-source, Xen virtualization system, and investigate the performance of the typical parallel programming paradigm MPI in PVC. The results of experiments show that the PVC is a helpful computing mode for a personal user in a LAN, and the application with MPI without much communication between different processes can achieve good performance in PVC.
    The 24th International Symposium on Computer and Information Sciences, ISCIS 2009, 14-16 September 2009, North Cyprus; 01/2009