Search

This paper proposes a novel context-based search method on edge computing for finding data of moving people. As the Internet of Things (IoT) spreads, many sensors are going to be connected to a wide area network. To utilize sensed data for various services, they first need to be searched for. Different from term-based web content search, sensed data require context to be derived by heavy load processing such as image analysis from enormous data. Therefore, it incurs a huge amount of time and cost. Time and cost can be reduced by estimating the edge server that stores the required data. This paper proposes a novel context-based search method that selects data to be analyzed on the basis of the existence probability of context in accordance with the installation area of edge servers and the data generation time. Also, it selects a server with low load to reduce response time. Since the proposed method incorporates not only low layer information such as computer load but also high layer information such as the existence probability of context in the data, the method enables context-based search for sensed data quickly with low cost. Simulations showed that the proposed method reduces the response time by 95% compared with selecting data without using the existence probability of context.

This paper proposes a data storage method that cost-efficiently manages a massive amount of short lifetime data continuously generated typically by IoT devices (called Live Data, e.g., surveillance camera images) considering real-time data retrieval. Along with the rapid growth of the IoT, many users expect to shift to an Open IoT in which they can share enormous amounts of Live Data. To efficiently use Live Data, it is important to use both hot and cold data storage. Data should be stored in cold storage, which is less expensive but has slow access speeds than hot storage, only if the data are unlikely to be used. The proposed method virtualizes distributed hot storage to provide two data placement mechanisms; (1) an original-data-only optimized allocation according to demand by migrating original data to the appropriate distributed hot storage, and (2) integrated data replacement in the case of capacity overflow by migrating data to the distributed hot storage storing less useful data. As a result of a performance evaluation, compared with a cache algorithm often used for traditional data, the proposed method can reduce operational costs (transmission costs for data migration and retrieval and miss–hit penalty costs) when the ratio of single miss–hit penalty costs to 1-hop transmission costs is more than 6.

This paper presents the performance evaluation of a metadata database (DB) management method that uses realistic numeric examples for IoT Live Data. The method is proposed to reduce the handling costs of Live Data. Live Data are here defined as data that are typically continuously generated by IoT devices and have short lifetimes (e.g., 10 fps surveillance camera images). We have already proposed an evaluation model in which the high locality is significantly featured in Live Data usage. The previous evaluation results are obtained only from general parameter values in statistical distributions. To evaluate realistic situations, this paper assumes that the metadata of Live Data with high usefulness for sharing by many users/services would dominate all metadata. In particular, for such data, we use both surveillance camera images and social networking service contents. The median values and the expected values are set considering the surveillance camera's locality (defined as the average distance between a surveillance camera and the users of its camera images). As a result, the proposed method can reduce the DB update costs by 99.0% while the additional search costs are reduced by up to 27.8% compared with the conventional metadata management method. The additional search costs are negligible compared with the reduction in DB update costs, since the number of searches is much smaller than the number of DB updates with respect to the number of update/search epochs.

This paper proposes a new Information-Centric Network architecture (ICN) suitable for handling metadata of the data which is continuously generated typically by IoT devices and/or has a short lifetime (called Live Data). The proposed ICN has two methods considering the feature of high locality of Live Data retrieval; (1) metadata of Live Data is stored for queries either on metadata databases (DB) or data generation terminals (publishers) with considering popularity and locality of the data, and a query is sent either to DB or publishers in a local area of the query issue terminals (subscribers). (2) the metadata DBs consist of one global DB and multiple local DBs which is for offloading the global DB and covers only local data. In the first method, we found that DB update cost reduced by 99.73(%) while suppressing search energy increase. Also, in the second method, we showed that DB update cost reduced by 85.70(%) while suppressing search energy increase.

In Information-Centric Network or location-free contents network, it is a big challenge to manage metadata of the data continuously generated with short lifetime (called Live Data), for example, surveillance camera data. This paper proposes a new Information-Centric Network architecture suitable for handling Live Data by considering its feature of high locality in data retrieval. The architecture includes methods of (1) having local metadata databases and categorizing data into global and local for the databases and (2) assigning MAC address of data storage in a content ID of the metadata. Quantitative evaluations revealed the both are effective enough to reduce read/write transactions in the metadata databases.

IoT technologies have been progressed. Now Open IoT concept has attracted attentions which achieves various IoT services by integrating horizontal separated devices and services. For Open IoT era, we have proposed the Tacit Computing technology to discover the devices with necessary data for users on demand and use them dynamically. However, existing Tacit Computing does not care about performance and operation cost. Therefore in this paper, we propose an automatic GPU offloading technology as an elementary technology of Tacit Computing which uses Genetic Algorithm to extract appropriate offload loop statements to improve performances. We evaluate a C/C++ image processing to verify effectiveness of GPU offloading and confirm more than 16 times performances within 1 hour calculation time.

A wide variety of devices is being installed in various environments such as homes, factories, and streets with the rapid expansion of the Internet of Things (IoT). To properly and securely use IoT devices, the states of various devices that have different properties and protocols must be managed. However, it is difficult to understand the consistency of an individual device when its installation place or software changes, since many individual IoT devices do not have unique identifiers. In this paper, we propose an architecture that estimates individual devices by analyzing and combining multiple pieces of information that can be obtained from the device. This architecture estimates individual device identity based on the time change pattern of the feature amount extracted from a signal transmitted by a device. Results of a simulation revealed that this architecture could identify the individual devices to find the correct device.

New services can be developed by combining some of the huge number of heterogeneous devices connected to a network in the era of the Internet of Things (IoT). However, to develop such services, a mechanism is required to detect devices with appropriate attributes such as functions or locations and to communicate with the devices. In this paper, we propose architecture that enables us to communicate with devices that have desired attributes in the IoT world by reducing the size of tables of devices' attributes by distributing the tables over the whole network. Moreover, we quantitatively describe how small the tables can be compared with centralized management at places such as datacenters and then discuss techniques and problems to make distributed tables a reality.

The number of physical resources, such as embedded sensors, actuators, and processors, that are connected to the Internet is increasing, leading to the formation of the Internet of Things. We argue that the discovery of adequate resources is difficult but important for open IoT services that use various IoT resources among various protocols and platforms. To discover appropriate resources for services among many resources, discovery considering live data is important. In contrast to current Web search engines, live data are dynamic, so it is difficult to adopt current architectures. In this paper, we discuss how live data search can be enabled with a distributed buffer mechanism and propose a live data search architecture for IoT.

This paper reports on new device for 3D cluster formation of stem cells under combinatorial seeding conditions - cell population and density. The device has an array of tapered holes of various sizes, and the surface is modified with hydrophilic polymer to prevent cell adhesion. Thus, the cells seeded onto the device aggregate in the areas defined by the holes. By varying the shape of the holes, cluster cultures can be performed under different conditions at one time. The device is easily incorporated into conventional cluster culture protocols by just placing the device onto an arbitrary culture substrate. Combinatorial cluster formation of stem cells derived from pancreatic duct was successfully demonstrated on ultra low attachment surface and Matrigel. The device should be useful for one-step combinatorial analysis to find optimal cluster culture conditions for various types of cells, and the more, for mass production of standardized embryonic body (EB) necessary in tissue engineering research.