The fast development of biomolecular structure determination has enabled the fine-grained study of objects in the micro-world, such as proteins and RNAs. The world is benefited. However, as the computational algorithms are constantly developed, the enrichment of features increases the algorithmic complexity and brings more computationally unfriendly modules. It calls for efficient solutions to leverage the rich and various hardware resources from the world’s most state-of-the-art supercomputing systems, and to fully accelerate the performance of the applications. In this article, we present our efforts on porting and optimizing the 3D reconstruction of RELION, one of the most popular cryo-EM software for biomolecular structure determinations, by leveraging different resources of the latest generation of Sunway heterogeneous supercomputer. Several novel approaches are proposed to resolve different challenges faced by the complex algorithm, including a multi-level parallel scheme and operator optimizations to smartly map and scale RELION, efficient strategies to largely address the memory bottlenecks and improve data locality, lock-free writing solutions to minimize write-write conflicts, and pipelining approaches to obtain excellent computation and communication overlap. Combining all proposed optimizations, the computation time is greatly reduced to under 2 hours, achieving 11.9× and 8.9× speedups on two different datasets. The overall design scales to 131,072 cores, increasing parallel efficiency from 33% to 61% and from 46% to 70%, respectively. To the best of our knowledge, this is the first work that fully optimized and scaled the 3D reconstruction of RELION using the latest Sunway system.

As the amount of data and computation of workflow applications continue to grow, the inter-cloud environment provide with a great amount of computing resources to respond to the increasing demand, but how to effectively map workflow sub tasks to cloud service providers for better QoS(quality of service) performance is a question. Remote sensing applications have massive data, frequent and large data transfers across the cloud could cause huge performance bottlenecks, will seriously affect the result of QoS such as makespan and cost. By using DTG(data transformation graph) model to study the data transmission process of global drought monitoring application, an optimized scientific workflow scheduling based on genetic algorithm is proposed for inter-cloud environment, experimental results show that this method can significantly optimize makespan and cost for data-intensive applications like remote sensing application and can effectively reduce the impact of performance bottlenecks.

BLAST[1] (Basic Local Alignment Search Tool) is a suite of programs used to identify similarity between genetic sequences. It is one of the most widely used tools in Bioinformatics. In recent years, with the size of gene and protein sequence database increasing exponentially, BLAST has become both a data-intensive and a computation-intensive application. How to run BLAST rapidly with low cost has always been the hotspot to researchers. Parallelization is one of the most important ways to resolve this problem. In this paper, a new approach for parallelizing BLAST based on a parallel processing framework called Robinia is presented. Compared with parallel version of BLAST presented before, Robinia-based BLAST has easy public accessibility and good scalability. Most importantly, it can support operation on WAN, this make it possible to integrate computation and storage resources on Internet to service for super-large scale BLAST projects. We implemented the Robinia-based BLAST and experimented on it using two different datasets. The results show that parallel BLAST based on Robinia can achieve linear speedup based on number of used nodes with good scalability and low cost.

Because the complexity of grid architecture and programming interface, it is difficult for beginners to develop new applications in spatial information grid (shortly SIG). This holds SIG back from infrastructure of geo-computation. To make the application development simpler, a situation driven application generation model is proposed in this paper. First, situation is introduced to describe application requirements and available resources. Based on the study on definition, acquirement, processing and matching of situation, the situation driven grid application generation model is proposed, which can create grid applications for spatial information grid more simply and quickly. Experiments on the implementation and test show that situation driven application generation can create situation applications from some keywords or phrases inputted by users based on situation processing.

Data distribution is one of the most important steps in the remote sensing applications; it is the bridge between consumers and providers of remote sensing data. Most of the conventional data distribution systems provide Web-based user interface and support a small quantity of data types only. It is not friendly for integrated remote sensing applications. So, a distributed and platform-independent remote sensing data distribution framework which supports multi data type, multi data source, and machine-to-machine intelligent integration is needed. Based on grid technology, a 3-layered remote sensing data distribution system named SIG-DDS is designed and implemented in this paper. In order to make the system simple, extensible and platform-independent, a well-defined extensible data accessing language based on XML is adopted as the accessing language of data sources and data brokers. The distribution system can distribute different sources, types and formats of remote sensing data in one system and make it easier to build up a distributed integrated remote sensing application

Grid is focused by the researchers theses years. In grid, resource registry meta-service assumes the responsibility for the register managing and indexing of resources in grid. It's very important for the running of grid. In this paper, we present a resource description system and provide its reference by XML schema first. Then, we design the distributed resource registry meta-service and implement it based on Java language. At last, the test and tryout indicate that it can provide high-performance and robust resource registering and indexing services.

It is very important for the improvement of web service to develop high performance ondemand service by script language easily. In this paper, by the analyzing of on-demand service requirement, the architecture of on-demand service framework by script language is purposed first. Than, the framework was filled by Groovy script language as a sample, and named Groovy service. Furthermore, it is designed and implemented based on Java language and Tomcat/Axis. At last, the test shows that the script language based on-demand service architecture is more flexible and powerful without heavy overhead.