Tomoo Aoyama

Performance Evaluation on Server Consolidation Using Virtual Machines

A virtual machine (VM) is a logical machine having almost the same architecture as a real host machine, running an operating system (OS) in it. This is called full visualization where multiple OSs can run as is in the real host. Many current industries tend to use many virtual servers, which are servers running on VMs, consolidating many physical servers into a single or fewer real machines, resulting in higher resource utilization and smaller space consumption. This is called the server consolidation using VMs. Noticing this tendency we evaluate the virtualization overhead by measuring the performance of the transaction systems running database servers on VMs. Our research objectives are to evaluate the VM overhead and the virtualization overhead for heavy database servers running in VMs. We have found that the total CPU utilization in the two VMs running database servers increases to 2.5 times that in the native system with the same hardware resources, that is, requiring each server native CPU utilization be less than 40%

Practical use of two kinds of Huge Page

On the multi-tasking system, the numbers of applications which execute concurrently are increasing and the memories which each application need are also increasing. The memories of a node are getting bigger and bigger to satisfy these demands. But TLB is very scarce as before. Under these circumstances, the uses of Huge Page reduce TLB misses by reducing the numbers of needed entries of TLB, consequently give applications which need large memories a good effect in regarding to execution speed. But it cause large useless unused memories. Then we proposed two methods which can use Huge Page while saving memories on IA-64.. One of them fill a fragments which result from the use of large Huge Page with smaller Huge Page, another makes it possible to use the second HugeTLBfs (HugeTLBfs2). As a result of experimentation, proposed methods show the same speed as the original HugeTLBfs and good results in regarding to memory requirements.

A photographic detection of yellow sand inflow

We proposed approaches to make visualizations of the yellow sand. One is the mod-pictures, which show SPM distribution structures in the atmosphere. Another is the RGB brightness, which shows the characters of the SPM. If we can photograph the mass of the SPM from a far point, we can find the 3-dimensional characters of the SPM. We believe that those approaches are useful to detect the yellow sand.

CPU vs. NIC one to one correspondence method for improving parallel processing performance in a small-way SMP cluster

We have studied various methods for improving the performance of SMP clusters. Most of them are using special hardware, non standard protocols, so that need special added library and complex programming, tending to raise their cost and to spoil their flexibility. To solve their problems we propose a CPU vs. NIC one to one correspondence method for improving parallel processing performance in a PC cluster, using comparatively simple method where transmitting NICs are fixedly assigned to CPUs one to one. So that the communication load between PCs can be balanced. Applying this method to parallel processing in a dual PC cluster, and comparing this method with conventional ordinary ones, we have obtained better result for some benchmarks.

Detection of local structure of yellow sand

The air pollution is the serious situation in East Asia. It is detected remarkably by the observations of suspended particulate matters. Yellow sand was observed conventionally in the district. On recent years, the yellow sand interacts with industrial waste gas and becomes polluted dust and goes over the border. We observed the yellow sand by using digital dust meters, and found that the yellow sand had local structures whose width was 6-75 km. This is narrower than that of a common sense. Passage times of the maximum parts of the peaks are less than 30 min. The maximum values are about 200 [mug/m3]. The yellow sand interacts with the vapor. The interactions are estimated by neural network analysis using optical SPM observations and meteorological measurements. We estimated some yellow sand peaks and showed different characters of each peaks.

Designing an XMLDB for an embedded system

Nowadays, large organizations are giving importance to XML, which is likely to become the data exchange standard for all organizations, irrespective of size. To manage large size XML data effectively, many XMLDB (XML database) management systems have been developed for desktops, laptops and servers. But these DBMSs are too large in size to fit into an embedded system. Usually, the embedded system has fewer resources compared to those desktops or laptops. Therefore, there is a necessity to build up a lightweight XMLDB so that it can be run even in an embedded system. This research is aimed to develop a high performance XMLDB for the embedded system. Using this XMLDB it will be possible to handle XML data efficiently and reduce the burden on an application developer. But, there is a problem of carrying out high-speed database processing in a resource constrained environment since common XMLDB is carried out on a high-performance machine. Here, we implemented a lightweight XMLDB with minimum functions as storing of XML data and a search of XML data. And as future work, we suggest to add more functions and methods for further improvement of the system.

Development of a multi-layer neural network for incomplete data set of environmental problems

Multi-layer neural networks are used for the multi regression analysis of many kinds of phenomena whose expressions are unknown. The application fields are environmental problems and medicine designs. Where, we often find incomplete parts in descriptors, which make precision of the analysis be lower. Moreover, the incomplete parts make the linked parts of other descriptors be invalid. We often cannot calculate multi regression analysis, therefore, we wish to eliminate the wrong effects. In the paper, we discuss some approaches to eliminate the wrong effects, and derive a method on neural networks, which compensates defect descriptors. We call the method compensation quantitative structure-activity relationships method (CQSAR).

Neural Network Expression for Water Purification in a River and the Application to Tamagawa in Tokyo

We discuss water purification in a river through cities, where indexes of the water quality are BOD, COD, T-N and T-P. To express changes of the quality, we introduce a model for purify functions in a river. The function is complex and non-linear. So, we use multi-layer neural networks and construct a function in the network, and use the derivatives for the cause and the effect of water quality. As for the learning and test data, we use ideal data derived from observations of the Tamagawa in Tokyo and add uniform random numbers to them. Since the model calculations represent river purification well, we apply the model to the Tamagawa. Thus, we can estimate the water purification functions, and find the river has no decomposing power for the nitrogen substances today

An Approach to Plant Identification Technology: Development of Lignin Pyrolysis Gas Chromatography and Pattern Recognition

We discovered a technique to extract a characteristic spectrum pattern from Lignin pyrolysis gas chromatography. The Lignin is a kind of methoxyphenol polymer that is often found among cell walls of plants. Using the spectrum, we can determine the species from fragments of 1 mg samples. The technique is useful for criminal investigation. The determination of species depends on distances calculated by species and a database of the spectra. We use the distances and can calculate family-locations in Plantae. The location shows the evolution/end of families. We evaluated 88 kinds of plants and discussed the end of ginkgo family and the branch of monocotyledon and a relation between C3/C4 metabolisms of Gramineae (family of the rice)