Pangfeng Liu

Energy-Aware Virtual Machine Dynamic Provision and Scheduling for Cloud Computing

Power consumption is one of the most critical problems in data centers. One effective way to reduce power consumption is to consolidate the hosting workloads and shut down physical machines which become idle after consolidation. Server consolidation is a NP-hard problem. In this paper, a new algorithms Dynamic Round-Robin (DRR), is proposed for energy-aware virtual machine scheduling and consolidation. We compare this strategy with the GREEDY, ROUNDROBIN and POWERSAVE scheduling strategies implemented in the Eucalyptus Cloud system. Our experiment results show that the Dynamic Round-Robin algorithm reduce a significant amount of power consumption compared with the three strategies in Eucalyptus.

Optimal placement of replicas in data grid environments with locality assurance

Data replication is a typical strategy for increasing access performance and data availability in data grid systems. Work on data replication in grid systems focuses on infrastructure for replication and mechanisms for creating/deleting replicas. The important problem of choosing suitable locations for placing replicas in data grids has not been well studied. In this paper, we address the problem of data replica placement in data grids given the traffic pattern and locality requirements. We propose a new placement algorithm that finds the optimal locations for the replicas so that the workload among these replicas is balanced. We also propose a new algorithm to decide the minimum number of replicas required when the maximum workload capacity of each replica server is known. All these algorithms ensure that locality requirements from the users are satisfied

HQEMU: a multi-threaded and retargetable dynamic binary translator on multicores

Dynamic binary translation (DBT) is a core technology to many important applications such as system virtualization, dynamic binary instrumentation and security. However, there are several factors that often impede its performance: (1) emulation overhead before translation; (2) translation and optimization overhead, and (3) translated code quality. On the dynamic binary translator itself, the issues also include its retargetability to support guest applications from different instruction-set architectures (ISAs) to host machines also with different ISAs, an important feature for system virtualization. In this work, we take advantage of the ubiquitous multicore platforms, using multithreaded approach to implement DBT. By running the translators and the dynamic binary optimizers on different threads on different cores, it could off-load the overhead caused by DBT on the target applications; thus, afford DBT of more sophisticated optimization techniques as well as the support of its retargetability. Using QEMU (a popular retargetable DBT for system virtualization) and LLVM (Low Level Virtual Machine) as our building blocks, we demonstrated in a multi-threaded DBT prototype, called HQEMU, that it could improve QEMU performance by a factor of 2.4X and 4X on the SPEC 2006 integer and floating point benchmarks for x86 to x86-64 emulations, respectively, i.e. it is only 2.5X and 2.1X slower than native execution of the same benchmarks on x86-64, as opposed to 6X and 8.4X slowdown on QEMU. For ARM to x86-64 emulation, HQEMU could gain a factor of 2.4X speedup over QEMU for the SPEC 2006 integer benchmarks.

Energy-efficient Virtual Machine Provision Algorithms for Cloud Systems

Power consumption is one of the most critical problems in data centers. One effective way to reduce power consumption is to consolidate the hosting workloads and shut down physical machines which become idle after consolidation. Server consolidation is a NP-hard problem. In this paper, we propose two new algorithms, Dynamic Round-Robin (DRR) and Hybrid, which combines DRR and First-Fit, for energy aware virtual machine scheduling and consolidation. We also propose an accurate power model to estimate the power consumption resulted from each algorithm. Strategies we proposed are compared with GREEDY, ROUNDROBIN and POWERSAVE scheduling strategies implemented in the Eucalyptus Cloud system. Our experiment results show that our DRR and Hybrid algorithms reduce power consumption by 56.4% and 55.9% respectively, compared with the ROUNDROBIN scheduling strategy in Eucalyptus. DDR and Hybrid also result in 3% less power consumption on average, compared with the POWERSAVE scheduling strategy in Eucalyptus.

An atomic model for message-passing

This paper presents a simple atomic model of message-passing net work syst ems. Within one synchronous time step each processor can receive one atomic message, perform local computation, and send one message. When several messages are destined to the same processor then one is transmitted and the rest are blocked. Blocked messages cannot be retrieved by their sending processors; each processor must wait for its blocked message to clear before sending more messages into the network. Depending on the traffic pattern, messages can remain blocked for arbitrarily long periods. The model is conservative when compared with exisiting message-passing systems. Nonetheless, we prove linear speedup for backtrack and branchand-bound searches using simple randomized algorithms,

Optimal replica placement strategy for hierarchical data grid systems

Grid computing is an important mechanism for utilizing distributed computing resources. These resources are distributed in different geographical locations, but are organized to provide an integrated service. In order to speed up data access efficiency data grid systems replicate essential data in multiple locations, so that a user can access the data from a site in his vicinity. This paper studies replica placement in data grid systems, taking into account several important issues described below. First, the replicas should be placed in proper server locations so that the workload on each server is balanced. Second, we choose the optimal number of replicas to balance the data access efficiency, and the expensive maintenance costs for multiple copies of data. Clearly, optimizing access cost of data requests and reducing the cost of replication are two conflicting goals. Finding a good balance between them is a challenging task. We propose efficient algorithms for selecting optimal locations for placing the replicas so that the workload among these replicas is balanced. Also when given the data usage from each user site and the maximum workload allowed for each replica server, our algorithm efficiently determines the minimum number of replicas required, as well as their locations.

Broadcast Scheduling Optimization for Heterogeneous Cluster Systems

Network of workstation (NOW) is a cost-effective alternative to massively parallel supercomputers. As commercially available off-the-shelf processors become cheaper and faster, it is now possible to build a PC or workstation cluster that provides high computing power within a limited budget. However, a cluster may consist of different types of processors, and this heterogeneity within a cluster complicates the design of efficient collective communication protocols. This paper shows that a simple heuristic called fastest-node-first (FNF) (1998, M. Banikazemi, V. Moorthy, and D. K. Panda, in “Proceedings of the International Parallel Processing Conference”) is very effective in reducing the broadcast time for heterogeneous cluster systems. Despite the fact that the FNF heuristic fails to give the optimal broadcast time for a general heterogeneous network of workstations, we prove that FNF always gives the optimal broadcast time in several special cases of clusters. Based on these special case results, we show that FNF is an approximation algorithm that guarantees a competitive ratio of 2. From these theoretical results we also derive techniques to speed up the branch-and-bound search for the optimal broadcast schedule in HNOW.

Optimal replica placement in hierarchical Data Grids with locality assurance

In this paper, we address three issues concerning data replica placement in hierarchical Data Grids that can be presented as tree structures. The first is how to ensure load balance among replicas. To achieve this, we propose a placement algorithm that finds the optimal locations for replicas so that their workload is balanced. The second issue is how to minimize the number of replicas. To solve this problem, we propose an algorithm that determines the minimum number of replicas required when the maximum workload capacity of each replica server is known. Finally, we address the issue of service quality by proposing a new model in which each request must be given a quality-of-service guarantee. We describe new algorithms that ensure both workload balance and quality of service simultaneously. We conduct extensive simulation experiments to evaluate the effectiveness of our algorithms. The comparison with the previous Affinity Replica Location Policy demonstrates that our algorithms consistently outperform the heuristic algorithm both in terms of minimum number of replicas used and the actual data transmission time.

SQLMR : A Scalable Database Management System for Cloud Computing

As the size of data set in cloud increases rapidly, how to process large amount of data efficiently has become a critical issue. MapReduce provides a framework for large data processing and is shown to be scalable and fault-tolerant on commondity machines. However, it has higher learning curve than SQL-like language and the codes are hard to maintain and reuse. On the other hand, traditional SQL-based data processing is familiar to user but is limited in scalability. In this paper, we propose a hybrid approach to fill the gap between SQL-based and MapReduce data processing. We develop a data management system for cloud, named SQLMR. SQLMR complies SQL-like queries to a sequence of MapReduce jobs. Existing SQL-based applications are compatible seamlessly with SQLMR and users can manage Tera to PataByte scale of data with SQL-like queries instead of writing MapReduce codes. We also devise a number of optimization techniques to improve the performance of SQLMR. The experiment results demonstrate both performance and scalability advantage of SQLMR compared to MySQL and two NoSQL data processing systems, Hive and HadoopDB.

Reduction optimization in heterogeneous cluster environments

Network of workstation (NOW) is a cost-effective alternative to massively parallel supercomputers. As commercially available off-the-shelf processors become cheaper and faster, it is now possible to build a cluster that provides high computing power within a limited budget. However, a cluster may consist of different types of processors and this heterogeneity complicates the design of efficient collective communication protocols. For example, it is a very hard combinatorial problem to find an optimal reduction schedule for such heterogeneous clusters. Nevertheless, we show that a simple technique called slowest-node-first (SNF) is very effective in designing efficient reduction protocols for heterogeneous clusters. First, we show that SNF is actually an approximation algorithm with competitive ratio two. In addition, we show that SNF does give the optimal reduction time when the cluster consists of two types of processors, anal the ratio of communication speed between them is at least two.