This paper studies five real-world data intensive workflow applications in the fields of natural language processing, astronomy image analysis, and web data analysis. Data intensive workflows are increasingly becoming important applications for cluster and Grid environments. They open new challenges to various components of workflow execution environments including job dispatchers, schedulers, file systems, and file staging tools. Their impacts on real workloads are largely unknown. Under- standing characteristics of real-world workflow applications is a required step to promote research in this area. To this end, we analyse real-world workflow applications focusing on their file access patterns and summarize their implications to schedulers and file system/staging designs.

This paper describes the rational behind designing workflow systems based on the Unix make by showing a number of idioms useful for workflows comprising many tasks. It also demonstrates a specific design and implementation of such a workflow system called GXP make. GXP make supports all the features of GNU make and extends its platforms from single node systems to clusters, clouds, supercomputers, and distributed systems. Interestingly, it is achieved by a very small code base that does not modify GNU make implementation at all. While being not ideal for performance, it achieved a useful performance and scalability of dispatching one million tasks in approximately 16,000 seconds (60 tasks per second, including dependence analysis) on an 8 core Intel Nehalem node. For real applications, recognition and classification of protein-protein interactions from biomedical texts on a supercomputer with more than 8,000 cores are described.

This paper studies five real-world data intensive workflow applications in the fields of natural language processing, astronomy image analysis, and web data analysis. Data intensive workflows are increasingly becoming important applications for cluster and Grid environments. They open new challenges to various components of workflow execution environments including job dispatchers, schedulers, file systems, and file staging tools. The keys to achieving high performance are efficient data sharing among executing hosts and locality-aware scheduling that reduces the amount of data transfer. While much work has been done on scheduling workflows, many of them use synthetic or random workload. As such, their impacts on real workloads are largely unknown. Understanding characteristics of real-world workflow applications is a required step to promote research in this area. To this end, we analyse real-world workflow applications focusing on their file access patterns and summarize their implications to schedulers and file system/staging designs.

Distributing large data to many nodes, known as a broadcast or a multicast, is an important operation in parallel and distributed computing. Most previous broadcast algorithms explicitly or implicitly try to deliver data to all nodes in the same rate. This assumption is reasonable for homogeneous environments where all nodes have similar receiving capabilities. However, when nodes have various receiving capabilities, nodes with slow-receiving capabilities slow down the entire receiving bandwidth in these algorithms. In such settings, each node desires to receive data at its largest possible bandwidth and to start computation as soon as it receives the data. In this paper, we propose to say a broadcast is stable when the bandwidth to a node is never sacrificed by the presence of other, possibly slow, receiving nodes, and proposes the stability as a desired property of broadcast algorithms. In addition, we show a simple and efficient stable broadcast when the topology among nodes is a tree and each link has a symmetric bandwidth. This work improves upon previously proposed algorithms such as FPFR and Balanced Multicasting. For general graphs, it outperforms them when the network is heterogeneous and for trees, our algorithm is proved to be stable and optimal. In a real environment with 100 machines in 4 clusters, our scheme achieved 2.1 to 2.6 times aggregate bandwidth compared to the best result in the other algorithms. We also demonstrated the stability by adding a slow node to a broadcast. Some simulations also showed that our algorithm also performs well in many bandwidth distributions.

The challenge of speech modification is to flexibly modify the speech without degrading speech quality. The conventional methods are limited by their inability to flexibly control speech signals in time and frequency domains. This causes degradationof the quality of modified speech. This paper proposes a highquality analysis/synthesis method for speech modification. To control the temporal evolution, we use a speech analysis techniquecalled temporal decomposition (TD), which decomposes a speech signal into event targets and event functions. The same event functions evaluated for the spectral parameters are alsoused to model the temporal evolution of the excitation parameters. The event functions describe the temporal evolution of the spectral and excitation parameters, and the event targets represent the “ideal” spectral parameters. To flexibly control speech signals in both time and frequency domains, we propose new methods to model the event functions and the event targets. The experimental results show that our proposed analysis/synthesis method produces high-quality synthesized speech, and allows the flexibility to modify speech signals.