Members Only :

Grid

The Grid group is conducting research and development on latest optimization techniques and parallel algorithms to carry out High Performance Computing Cluster (HPCC).

Research is engaged in utilizing latest algorithms in order to optimize the computing performance. By using the latest algorithms, Grid group could solve the graph theory problems. In addition, Grid group could develop hybrid algorithms to solve high computing performance problems.Grid group also have enhanced the computing performance by using parallel computing.

Demonstration of research

In addition, many of these optimization problems are difficult to solve, which belong to NP hard or NP complete class. As such a problem is difficult to find the optimum solution, we generally find the approximate solution, which is not the optimum solution but a solution close to the optimum solution.
  • Traveling Salesman Problem(TSP)

    The Travelling Salesman Problem (often called TSP) is a classic algorithmic problem in the field of combinatorial optimization. "Given cities and the distances between cities, find the shortest route that visits each city exactly once and returns to the origin city."


  • Integer Programming(IP)

    IP is also called integer programming, and belongs to the field of mathematical optimization. Objective function and constraints are defined by linear inequalities, equality in the same manner as the Linear Programming (LP), but to the LP in the solution space is continuous, IP is given as a discrete space. Whereas LP can be solved in polynomial time, IP cannot be solved.

    etc...

Demonstration of Algorithm development

The following research is carried out at a well-equipped development environment with abundant computing resources.
  • Implementation of latest algorithm

    The algorithms developed by other researchers have not been used to solve the TSP problem yet. Grid group modifies the algorithms in order to use to solve the TSP problems. By utilizing the advantages of developed algorithms such as highly scalable, high-speed operation and low computational cost, a TSP solver to effectively solve the TSP is developing.


  • Hybrid algorithm

    A hybrid algorithm is a combination of two or more other algorithms that solve the same problem, either choosing one (depending on the data), or switching between them over the course of the algorithm. It is addressing the main problems such as accuracy of the result and time using to process it. For example, there is an algorithm which produces high accuracy results but takes long time to process, in other hand there is an algorithm which produces not very good accuracy results and take short time to process. Hybrid algorithm is able to address this problem and produce high accuracy results with short time.


  • Massively parallel computing

    In computing, massively parallel computing refers to the use of a large number of processors to perform a set of coordinated computations in parallel. The developing algorithm tries to improve the performance of parallel computing in an advanced way to utilize the maximum hardware capacity of the computer. Main features of this algorithm reduce computing cost and keep the waiting time near zero; therefore this is considered as one of the very challenging research in our group.

    etc...

Demonstration of research and development environment

  • - Intel Core-i CPU Series + Desktop with Linux(Ubuntu、Fedora、CentOS、etc...)

    Improving the development efficiency by using the latest processors based PCs and by operating the graphical integrated development environment (IDE) on powerful PC.


  • - Multi-Display

    Monitoring source codes and references simultaneously with dual displays.


  • - Integrated Development Environment (IDE) + Programming languages

    An integrated development environment (IDE) is a software application that provides comprehensive facilities to computer programmers for software development. It makes use of various integrated development environment in order to use different languages​. Such as, if you use C++ you can use IDE as Qt Creator, if you use Java you can use IDE as Eclipse, if you use R you can use IDE as R Studio.


  • - Co-processor

    Since the Infrastructure has able to support massively parallel algorithm, the Grid team is using the Xeon Phi co-processor. It has performance of 1 Tera FLOPS, 200 over threads, and memory bandwidth with 200-300 GB/s.


  • - HPC cluster

    Since the infrastructure to support massively parallel algorithm, we can also use the HPC cluster with low communication speed comparing to the co-processor.

Return to TOP

2015-M1 Kansai-u AL-LAB Ryouta Nagatsuji
"Kansai University, Algorithm Laboratory, description of GR.Grid"


    Study


Research activities in AL-lab are divided mainly into three groups:

Grid

Network

Web