Server Load Balancing (SLB) is a popular technique to build high-availability web services as offered from Google and Amazon for example. Credit based load balancing strategies have been proposed in the literature whe...
详细信息
Server Load Balancing (SLB) is a popular technique to build high-availability web services as offered from Google and Amazon for example. Credit based load balancing strategies have been proposed in the literature where the back end servers dynamically report a metric called Credit to the Load Balancer (LB) which reflects their current capacity. This enables the LB to adapt the load balancing strategy. The benefit of Credit based SLB has been shown by simulations, but up to now, it is not used in productive systems, since efficient implementations were missing. This paper presents the evaluation of an implementation of Credit based SLB, the so-called Self-Adapting Load Balancing Network (salbnet). We evaluate salbnet for a cluster of web servers. The measurements are done with a representative workload based on a Wikipedia trace and confirm the benefit of the self-adapting load balancing approach.
Efficient sorting is vital for overall performance of the underlying application. This paper presents Butterfly Network Sort (BNS) for sorting large data sets. A minimal version of the algorithm Min-Max Butterfly is a...
详细信息
ISBN:
(纸本)9780889869431
Efficient sorting is vital for overall performance of the underlying application. This paper presents Butterfly Network Sort (BNS) for sorting large data sets. A minimal version of the algorithm Min-Max Butterfly is also shown for searching minimum and maximum values in data. Both algorithms are implemented on GPUs using OpenCL exploiting data parallelism model. Results obtained on different GPU architectures show better performance of butterfly sorting in terms of sorting time and rate. The comparison of butterfly sorting with other algorithms:bitonic, odd-even and rank sort show significant speedup improvements against all on Nvidia Quadro-6000 GPU with relatively better sorting time and rate.
In this paper we present a data model view on the MEDIOGRID project. Based on the general project architecture, we define the data model requirements and technology dependencies. We introduce a two layer data manageme...
详细信息
ISBN:
(纸本)9780889866379
In this paper we present a data model view on the MEDIOGRID project. Based on the general project architecture, we define the data model requirements and technology dependencies. We introduce a two layer data management architecture based on highly interoperable web service interfaces and describe computational orchestration. We suggest a resource centric view and highlight improvements at the scheduling level based on the proposed data model view.
The idea behind Cloud computing is to deliver Infrastructure-, Platform-, and Software as a Service (IaaS, PaaS, and SaaS) on a simple pay-per-use basis. In this paper, we introduce our work, OSGi Service Platform as ...
详细信息
ISBN:
(纸本)9780889868649
The idea behind Cloud computing is to deliver Infrastructure-, Platform-, and Software as a Service (IaaS, PaaS, and SaaS) on a simple pay-per-use basis. In this paper, we introduce our work, OSGi Service Platform as a Service (OSPaaS), a PaaS model for running an OSGi service platform in the cloud for e-Learning and teaching purposes. OSPaaS leverages OpenNebula, a virtual infrastructure manager, to dynamically launch virtual machines (VMs) on idle resources or dedicated servers. In addition, OSPaaS uses Shibboleth as a Single Sign-On mechanism for seamless authentication and authorization. To assess the suitability of OSGi for cloud computing, this paper investigates and analyzes three OSGi frameworks, i.e. Knopflerfish, Equinox and Apache Felix. Subsequently, an OSPaaS architecture is presented and described. Finally, this paper shows a use case scenario and advantages of OSPaaS for e-Learning & teaching purposes.
In the modern business world, very often two parties collaborate with each other for their mutual benefit. Accordingly, transactions are processed between them, with one party processing a part of a transaction, and t...
详细信息
ISBN:
(纸本)9780889867840
In the modern business world, very often two parties collaborate with each other for their mutual benefit. Accordingly, transactions are processed between them, with one party processing a part of a transaction, and the other continuing with the remainder. Because of the mutual advantage it brings to the collaborators, to discover such a pattern relation becomes especially important that, a frequent pattern of one party is dependent upon, or associated with, a frequent pattern of the other party. Generally it is required that pattern relation mining should be conducted without disclosing private data to each other. And also, since any subpattern of a frequent pattern is also frequent, it is sufficient to mine only the maximal frequent patterns. In this paper we propose an effective privacy-preserving maximal pattern relation mining algorithm, called CMPRM.
We investigate the relative significance of kernelization versus branching for parallel FPT implementations. Using the well-known vertex cover problem as a familiar example, we build and experiment with a testbed of f...
详细信息
ISBN:
(纸本)9780889869431
We investigate the relative significance of kernelization versus branching for parallel FPT implementations. Using the well-known vertex cover problem as a familiar example, we build and experiment with a testbed of five different classes of difficult graphs. For some, we find that kernelization alone obviates the need for parallelism. For others, we show that kernelization and branching work in synergy to produce efficient implementations. And yet for others, kernelization fails completely, leaving branching to solve the entire problem. Structural graph properties are studied in an effort to explicate this trichotomy. The NP-completeness of vertex cover makes scalability an extreme challenge. We mainly employ Hopper, named after the famous computing pioneer Admiral Grace Murray Hopper. The Hopper platform is currently one of the world's fastest supercomputers.
Hardware/software co-design for future-generation highperformance computing (HPC) systems aims at closing the gap between the peak capabilities of the hardware and the performance realized by applications (application...
详细信息
ISBN:
(纸本)9780889869431
Hardware/software co-design for future-generation highperformance computing (HPC) systems aims at closing the gap between the peak capabilities of the hardware and the performance realized by applications (applicationarchitecture performance gap). Performance profiling of architectures and applications is a crucial part of this iterative process. The work in this paper focuses on operating system (OS) noise as an additional factor to be considered for co-design. It represents the first step in including OS noise in HPC hardware/software co-design by adding a noise injection feature to an existing simulation-based co-design toolkit. It reuses an existing abstraction for OS noise with frequency (periodic recurrence) and period (duration of each occurrence) to enhance the processor model of the Extreme-scale Simulator (xSim) with synchronized and random OS noise simulation. The results demonstrate this capability by evaluating the impact of OS noise on MPI Bcast() and MPI Reduce() in a simulated futuregeneration HPC system with 2,097,152 compute nodes.
This paper proposes an optimization method of task-allocation for reducing contentions. There have been some attempts for optimizing task-allocation that minimizes the product of the amount of communications and the n...
详细信息
ISBN:
(纸本)9780889868649
This paper proposes an optimization method of task-allocation for reducing contentions. There have been some attempts for optimizing task-allocation that minimizes the product of the amount of communications and the number of the communication hops. However, since those methods do not consider the occurrence of contentions, the effect has not been sufficient. The method proposed in this paper uses information of concurrent communication to estimate the effect of contentions to find the optimal task-allocation. In three environments examined on the experiments, the proposed method has shown a better effect than the existing method on two environments, tree and fat tree. On these environments, the maximum gain of performance over the existing method was about 25%. On the other hand, on a mesh environment, IBM BlueGene/L, the existing method better effect than the proposed method. As one of the reasons for this, the influence of the packet priority on the network of BlueGene/L to the behavior of the proposed method is discussed.
With the richness of present-day hardware architectures, research effort has been going into tightening the revealed synergy between hardware and software. A large focus has been put on the creation of software tools ...
详细信息
ISBN:
(纸本)9780889867840
With the richness of present-day hardware architectures, research effort has been going into tightening the revealed synergy between hardware and software. A large focus has been put on the creation of software tools to facilitate hardware design. Moreover, enormous efforts have been invested to develop high-level methodologies, formal techniques, parallelization procedures, and synthesis tools that target state-of-the-art hardware architectures including Field-programmable Gate Arrays (FPGAs). In this paper, we explore the effectiveness of a formal methodology in the design of parallel versions of the current Advanced Encryption Standard {AES), namely, the Rijndael cryptographic algorithm. The suggested methodology adopts a functional programming notation for specifying algorithms and for reasoning about them. The parallel behavior of the specification is then derived and mapped onto hardware. Several parallel AES implementations are developed with different performance characteristics. The refined designs are tested under Celoxica's RC-1000 reconfigurable computer with its 2 million gates Virtex-E FPGA. Performance analysis and evaluation of the proposed implementations are included.
Analysis of existing research work indicates that preference for implementation of queries to structured data is given to parallel DBMS. MapReduce (MR) is perceived as supplementary to DBMS technology. We attempt to f...
详细信息
Analysis of existing research work indicates that preference for implementation of queries to structured data is given to parallel DBMS. MapReduce (MR) is perceived as supplementary to DBMS technology. We attempt to figure out behavior pattern of parallel rowstorage DBMS and MR system Hadoop on the example of Join task depending on the variation of the parameters that in other authors' experiments do not vary or differ from ours. This article presents detailed process models for table joins in the parallel row-storage DBMS and MRsystem, as well as the results of detailed calculation experiments performed on these models. The models were set up for various scalability schemes for MR (number of nodes) and DMBS (data volume in a node) and fragmentation of the joined tables by the primary key. The following parameters were varied: queried data selectivity, number of sorted resulting records and cardinality of the grouping attribute. The modeling results showed that with the increase of the stored data volume parallel DBMS starts losing against MR-system at certain thresholds.
暂无评论