FogGIS: Fog Computing for Geospatial Big Data Analytics

FogGIS: Fog Computing for Geospatial Big Data Analytics Rabindra K. Barik1, Harishchandra Dubey2, Arun B. Samaddar3, Rajan D. Gupta4, Prakash K. Ray5 1 School of Computer Application, KIIT University, Bhubaneswar, India rabindra.mnnit@gmail.com 2 Electrical Engineering,The University of Texas at Dallas, USA harishchandra.dubey@utdallas.edu 3 Director, NIT Sikkim, India absamaddar@yahoo.com 4 Civil Engineering Department, MNNIT Allahabad, India gupta.rd@gmail.com 5 Electrical Engineering Department, IIIT Bhubaneswar, India prakash@iiit-bh.ac.in Abstract— Cloud Geographic Information Systems (GIS) has emerged as a tool for analysis, processing and transmission of geospatial data. The Fog computing is a paradigm where Fog devices help to increase throughput and reduce latency at the edge of the client. This paper developed a Fog-based framework named Fog GIS for mining analytics from geospatial data. We built a prototype using Intel Edison, an embedded microprocessor. We validated the FogGIS by doing preliminary analysis. including compression, and overlay analysis. Results showed that Fog computing hold a great promise for analysis of geospatial data. We used several open source compression techniques for reducing the transmission to the cloud. Keywords—Cloud GIS;Compression; Geosptial Big Data;Overlay Analysis Fog Computing; I. INTRODUCTION Geographic Information System (GIS) is a system of software and computer hardware that enables end-users to retrieve, store, and analyze huge amount of geospatial data from a various sources [1]. GIS is applied in decision making, storage of various kinds of data, bringing data and maps to a common scale as per the user needs, superimposing, querying and analyzing the data and designing/ presenting final maps/ reports to the administrators and planners [2]. The utility of GIS for planning of land resources and decision making has become widely popular and are being used for a wide range of applications. GIS has emerged as a powerful tool in integrating and analyzing various thematic layers along with their attribute information to create and visualize alternative planning scenarios for planners and decision makers. The user friendliness of GIS is a feature that has made GIS a preferred platform for planning all over the world, coupled with various analysis and modelling functionalities. GIS can play an important role in various applications such as environmental monitoring, natural resource management, healthcare, land use planning and urban planning. GIS integrates common database operations such as query formation, statistical computations and overlay analysis with unique visualization and geographical functionalities. These characteristics distinguish GIS from other information systems and make it valuable to a wide range of public and private enterprises for explaining events, predicting outcomes and designing strategies. The GIS technology and cloud computing has been merged to perform a value added services that give rise to geospatial cloud computing. The geospatial data have rich information about temporal as well as spatial distributions. In traditional setup, we send the data to the cloud where these are going for further processing and analysis. The Fog Computing provides low-power gateway that can increase throughput and reduces latency near the edge of the geo-spatial clients. It reduces the storage needed for geospatial big data in the cloud. In addition, reduction in the required transmission power results in overall improvement in efficiency. Fog devices can act as a gateway between clients such as mobile phones [19]. In this paper, we let the geospatial data be processed at the edge using Fog computing device. The present paper made the following contributions to the GIS systems: • FogGIS framework is proposed for improved throughput and reduced latency for analysis and transmission of geospatial data • Intel Edison was employed as the fog computing device • Various compression techniques were used for reducing the data size, thereby reducing transmission power • Geospatial data analysis scheme, overlay analysis in thin clients environment was performed using FogGIS framework. We performed a case study by doing overlay analysis of city of Alaska, USA II. RELATED WORKS A. Geospatial Cloud Cloud computing provides adequate storage and computational infrastructure for implementation of geo-spatial analysis prototypes. This model provides a transition from PC to cloud servers. Cloud computing and other web processing architectures creates an open environment in web with shared assets [5-7]. Mobile Clients Thick Clients Client Tier Layer Thin Clients HTTP Request HTTP Response Application Server Application Tier Layer Catalog Services Data services Processing services CSW WMS WFS WCS WPS Data Tier Layer Data Providers Database PostGIS PostgreSQL Metadata File System Fig. 1. System architecture for Geospatial Cloud Model adapted from[10]. Geospatial Cloud delivers a platform in which organizations interrelate with technologies, tools and expertise to nurture deeds for producing, handling and using geographical statistics and data. Likewise, Geospatial Cloud deploy a unique-instance, multitenant design and permitting more than one client to contribute assets without disrupting each other. This integrated hosted service method helps installing patches and application advancements for user’s transparency. Its another characteristic is embrace of web services and as an established architectural methodology in engineering [8-9]. Many cloud platforms uncover the applications statistics and functionalities via web service. This permit clients to query/update different types of cloud services and applications data programmatically, along with the provision of a typical tool to assimilate different cloud applications in the software cloud with enterprise SOA infrastructure. Figure 1 shows the system architecture for Geospatial Cloud Model adapted from [10]. The client tier layer consists of thick clients, thin clients and mobile clients with visualization functionality for spatial information. Mobile clients are users operating through mobile devices. The users those are working on web browsers are defined to be thin clients. In thin clients, users do not require any additional software for the operation. Thick clients are the users processing or visualising the spatial data in standalone system where it requires additional software for full phase operation. The Application Tier comprises the main geo-spatial services executed by servers. It enables intermediate amongst the different clients and providers. In top of the application tier, dedicated server for application has been operated for different services i.e. Web Map Service (WMS), Web Coverage Service (WCS), Web Feature Service (WFS), Web Catalog Service (CSW) and Web Processing Service (WPS). The dedicated application server is responsible for requests to and response from client to application server. In addition, application services include three types of server application i.e. catalog servers, data servers and processing servers. Catalog severs are used to search the metadata information regarding the stored spatial data. Catalog server is one of the important system components for controlling spatial information in cloud environment. In the catalog service, a standard publish-find-bind service framework are implemented which has been defined by OGC web service architecture. Data server deals with the WMS, WCS and WFS [11]. Processing server offers a geospatial processes which allows different clients to smear in WPS standard spatial data. The detail explanation of every processes done by client request, forward the desire processing service with input of several factors, specifies and provides definite region in leaping box and feedbacks with composite standards. Data tier Layer comprises of the various data in spatial form and related info. System utilizes the layer to store, recover, manipulate and update the spatial data for further analysis. Data providers can be store in different open source DBMS packages, simple file system or international organizations (e.g., Bhuvan, USGS).It has been shown from the system architecture of Geospatial Cloud that geospatial data are one of the key components in data layer for the handling of huge amount of data in terms of various spatial analysis. The amount of data which has been handling in Geospatial Cloud computing, it requires geospatial data from the various components. That gives rise to the concept of geospatial big data aspects and that will discuss in the next section. B. Geospatial Big data Big data are data whose scale, distribution, diversity, and/or timeliness require the use of new technical architectures and analytics to enable insights that unlock new sources of business value. Big data usually includes data sets with sizes beyond the ability of commonly used software tools to capture, curate, manage, and process data within a tolerable elapsed time [12]. Big data can come in multiple forms. Most of the big data is semi-structured, Quasi structured or unstructured, which requires different techniques and tools to process and analyze. Analysis of data sets can find new correlations to "spot business trends, prevent diseases, combat crime and so on. Data sets are growing rapidly in part because they are increasingly gathered by cheap and numerous information-sensing mobile devices, aerial (remote sensing), software logs, cameras, microphones, vector data. Graph data appear in the form of road networks. Here, an edge represents a road segment and a node represents an intersection or a landmark. There are various regions behind the disadvantageous of geospatial cloud computing with geospatial big data. As we know reliability, manageability and cost saving, are the key factors in which cloud computing always be one of advantageous over other emerge technology for data processing. But in terms of security and privacy are the main concerns for the processing of sensitive data. Particularly in health geoinformatics scenario, data are so sensitivity for further processing and analysis [14]. Thus, for minimization of privacy and security risks, it has to be used as per the user Fig. 2. Conceptual diagram of the proposed FogGIS framework for power-efficient, low latency and high throughput analysis of the geospatial big data. radio-frequency identification (RFID) readers and wireless sensor networks. Geospatial data has always been big data with the combination of Remote Sensing, GIS and GPS data[13]. In these days, big data analytics for geospatial data is receiving considerable attention to allow users to analyze huge amounts of geospatial data. Geospatial big data typically refers to spatial data sets exceeding capacity of current computing systems. context for limited amount of data access within the limited framework. After processing within the limited framework, it will transfer to the next level for the final processing of data analysis. That wills benefits for data security and privacy. Thus, fog computing comes into picture for geospatial big data processing in our present study. Generally, geospatial data has been categorized into raster data, vector data and graph data. Raster data include geospatial images which are obtained by satellites, security cameras and aerial vehicles. The raster data has been provided by different government agencies for using in various analysis. It can be extract number of feature from these raster data. Change detection and pattern mining are two examples in which data analyst does. Vector data consist of points, lines and polygons features. For examples, in Google map, the various temples, bus stops and churches has been marked thorough points data whereas lines and polygons corresponds to the road networks. Spatial correction pattern analysis and hot spot detection are the analysis which can be done through C. Fog Computing Fog computing was coined by Cisco in 2012 [15]. It refers to a computing paradigm that uses interface kept close to the devices that acquire data. It introduces the facility of local processing leading to reduction in data size, lower latency, high throughput and high power efficiency of the cloud-based systems. It has been successfully used in smart cities [16] and healthcare [17]. The Fog devices are embedded computers such as Intel Edison that acts a gateway between cloud and mobile devices such as smart phones and mobile GIS. III. FOGGIS FRAMEWORK This section describes various components of the proposed FogGIS framework and discusses the methods implemented in it. We discuss the hardware, software and methods used for compression of geo-spatial big data. B. Lossless Compression Techniques In the present study, we have a number of popular compression algorithms for reducing the data size in fog layer. The concept of compression in GIS is not new, it have been used in network GIS and mobile GIS[23-25]. In this paper, we translated the compression from mobile GIS to fog layer [20]. The geo-spatial data is compressed on the Fog computer that later transmits the data to cloud layer. The cloud layer have the choice to store the compressed data or decompress it Fig. 3. Overlay operation on thick client environment in FogGIS framework. Fig. 4. Overlay operation on thin client environment in FogGIS framework. A. Intel Edison We employed Intel Edision as Fog computing device in proposed FogGIS framework [18]. Intel Edison is powered by a rechargeable lithium battery. It contains dual-core, dualthreaded 500MHz Intel Atom CPU along with a 100MHz Intel Quark microcontroller. It possess 1GB memory with 4GB flash storage. It supports IEEE 802.11 a,b,g,n standards and can connect to WIFI. We used UbiLinux operating system for running compression utilities. Figure 2 shows the proposed FogGIS framework. The fog device acts as a gateway between thick, thin and mobile clients and cloud layer. The proposed FogGIS framework has three layers as client tier layer, geospatial cloud layer and FogGIS layer. In client tier, the categories of users have been further divided into thick client, thin client and mobile client environment. Processing of geospatial data can be possible within these three environments. Geospatial Cloud layer is mainly focused on overall storage and analysis of geospatial data. The Fog layer works as middle tier between client tier layer and geospatial cloud layer. It has been experimentally validated that the fog layer is characterized by low power consumption, reduced storage requirement and overlay analysis capabilities. before processing, analysis and visualization. We used only lossless techniques in this paper such as .zip, .tar.gz, .gzip. The results have been obtained by using various lossless compression techniques done at the Fog gateway which has shown in Table I C. Geospatial Analysis of Alaska city, USA In this section, data analysis particularly overlay analysis is performed for city of Alaska, USA. Overlay Analysis is one of the important data analysis in which we superimpose various geospatial data in a common platform for better analysis of raster and vector geospatial data. We performed the case study on the city of Alaska, USA. We downloaded the freely available dataset both raster and vector geospatial data[20]. It has been found that one SRTM raster data and three number of vector data of Alaska in EPSG:2964 file format. Continents boundary, City boundary of Alaska and airport location details of Alaska are the three number of vector data have been used; which are in shape file formats. The overlay analysis of various vector data and raster data of particular area has been performed. Initially, the downloaded datasets have been opened with Quantum GIS; desktop based GIS analysis tools, and performed some join operations which has been shown in Figure 3. The desired overlay operation has been done with standalone application, are known as thick client operation. In Quantum GIS, plugin named as QGISCloud has been installed. The said plugin has the capability of storing various raster and vector data set in cloud database for further overlay analysis. After storing in cloud database, it also generates the mobile and thin client link for visualization of both vector and raster data set. Figure 4 shows the overlay operation on thin client environment. The Figure 3 and 4 shows the overlay analysis on thick and thin client respectively. We can see that the overlay analysis is a useful technique for visualization of geospatial data. TABLE I. PERFORMING COMPRESSION ON FOG GIS FRAMEWORK USING GLOBAL MAP DATA[26]. Geo-spatial Data Original .tar.gz .iso .zip .tar .gzip .zipx Data Size Compressed Compressed Compressed Compressed Compressed Compressed (in MBs) Size (in MB) Size Size Size Size Size (in MB) (in MB) (in MB) (in MB) (in MB) Line- 6.7 4.9 5.2 5.1 6.2 5.8 5.6 Coast LineGeodatabase 3.2 2.7 2.9 3.2 3.4 3.4 3.6 Political Boundaries Areas-Shapefile 47.3 33.7 33.6 33.4 32.8 Political Boundaries AreasGeodatabase 19.7 17.3 16.4 16.2 16.0 15.8 15.6 Political Boundaries LinesShapefile 47.5 19.6 24.5 25.2 24.6 24.4 26.2 Political Boundaries LinesGeodatabase 21 10.5 12.7 11.8 13.9 14.6 14.4 Canals AqueductsShapefile 2.1 1.1 1.2 1.5 1.7 1.8 1.9 Canals and AqueductsGeodatabase 1.5 0.932 0.942 0.938 0.936 0.939 0.942 Inland Water Areas-Shapefile 49.1 33.2 36.2 34.6 35.3 34.4 36.4 Inland Areas- 20.5 18.2 18.4 18.6 18.8 19.2 19.0 Water Courses— Shapefile 345.7 330.7 332.7 333.7 331.7 333.8 333.2 Water Courses— Geodatabase 163.9 105.1 111.9 110.4 110.6 110.2 111.8 Coast Shapefile and Water Geodatabase We used the global map data for benchmarking the various compression algorithms [26]. The Table I shows the compressed data size and original data size for various compression procedures. The compression procedures used are .tar.gz, .iso, .zip, .tar, .gzip, .zipx. Clearly, the compression ratio depends on the data type and size. However, the .tar.gz has consistently performed the best in terms of compression ratio for Global Map Data [26]. IV. CONCLUSIONS In this paper, we developed and validated FogGIS framework that employed Fog gateway in a cloud GIS model. Intel Edision processor was used as the fog computer. The Fog gateway reduces the storage space requirments, transmission power, increased throughput and reduced latency leading to overall efficiency of GIS system using FogGIS as an intermediate gateway. The. FogGIS framework introduces edge intelligence in geospatial cloud environment. In future, we would like to add more intelligent processing at the Fog layer in mobile client environments. REFERENCES [16]. [17]. [18]. [19]. [20]. [21]. [22]. [1]. Bonham-Carter, Graeme F, “Geographic information systems for geoscientists: modelling with GIS,” Elsevier, Vol. 13, 2014. [2]. Babiker, I.S., Mohamed, M.A., Terao, H., Kato, K. and Ohta, K., “Assessment of groundwater contamination by nitrate leaching from intensive vegetable cultivation using geographical information system,” Environment International, Vol. 29, No. 8, pp.1009-1017, 2004. [3]. http://maps.unomaha.edu/Peterson/gis/notes/GISAnal1.html [Accessed on 8th August, 2016] [4]. Peterson, M.P., “Mapping in the Cloud. Guilford Publications,” 2014. [5]. Buyya, R., Yeo, C.S. and Venugopal, S., “ Market-oriented cloud computing: Vision, hype, and reality for delivering it services as computing utilities,” 10th IEEE International Conference on High Performance Computing and Communications, pp. 5-13, 2008. [6]. Chen, Z., Chen, N., Yang, C. and Di, L., “Cloud computing enabled web processing service for earth observation data processing,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol. 5, No. 6, pp.1637-1649, 2012. [7]. Huang, Q., Yang, C., Liu, K., Xia, J., Xu, C., Li, J., Gui, Z., Sun, M. and Li, Z., “Evaluating open-source cloud computing solutions for geosciences,” Computers & Geosciences, Vol. 59, pp.41-52, 2013. [8]. Pandey, S., “Cloud Computing Technology & GIS Applications,” The 8th Asian Symposium on Geographic Information Systems From Computer & Engineering View (ASGIS 2010), ChongQing, China, pp. 1-2, 2010. [9]. Yang, Chaowei, Michael Goodchild, Qunying Huang, Doug Nebert, Robert Raskin, Yan Xu, Myra Bambacus, and Daniel Fay, “Spatial cloud computing: how can the geospatial sciences use and help shape cloud computing?,” International Journal of Digital Earth, Vol. 4, No. 4, pp.305-329, 2011. [10]. Evangelidis,K. , Ntouros,K., Makridis,S., and Papatheodorou, C., “Geospatial services in the cloud,” Computers & Geosciences, Vol. 63, pp. 116–122, 2014. [11]. Yue, P., Zhou, H., Gong, J. and Hu, L., “Geoprocessing in Cloud Computing platforms–a comparative analysis,” International Journal of Digital Earth, Vol. 6, No. 4, pp.404-425, 2013. [12]. Lee, Jae-Gil, and Minseo Kang, “Geospatial Big Data: Challenges and Opportunities,” Big Data Research , Vol. 2, No. 2, pp. 74–81, 2015. doi:10.1016/j.bdr.2015.01.003. [13]. Ma, Y., Wu, H., Wang, L., Huang, B., Ranjan, R., Zomaya, A. and Jie, W., “Remote sensing big data computing: challenges and opportunities,” Future Generation Computer Systems, Vol. 51, pp.47-60, 2015. [14]. Andreu-Perez, Javier, Carmen CY Poon, Robert D. Merrifield, Stephen TC Wong, and Guang-Zhong Yang, "Big data for health, " IEEE journal of biomedical and health informatics, Vol. 19, No. 4, pp.1193-1208, 2015. [15]. F. Bonomi, R. Milito, J. Zhu, and S. Addepalli, “Fog computing and itsrole in the internet of things,” in Proceedings of the first edition of theMCC workshop on Mobile cloud computing. ACM, pp. 13–16, 2012. G. P. Hancke, G. P. Hancke Jr et al., “The role of advanced sensing in smart cities,” Sensors, vol. 13, No. 1, pp. 393–425, 2012. H. Dubey, J. Yang, N. Constant, A. M. Amiri, Q. Yang, andK. Mankodiya, “Fog data: enhancing telehealth big data through fog computing,” in Proceedings of the ASE BigData & SocialInformatics2015. ACM, , pp. 14, 2015. V. Dastjerdi, H. Gupta, R. N. Calheiros, S. K. Ghosh, and R. Buyya, “Fog computing: Principals, architectures, and applications,” arXiv preprint arXiv:1601.02752, 2016. S. Yi, C. Li, and Q. Li, “A survey of fog computing: concepts, applications and issues,” in Proceedings of the 2015 Workshop onMobile Big Data. ACM, pp. 37–42, 2015. http://qgis.org/downloads/data/ [Access on 9th july 2016] http://qgiscloud.com/rabindrakumarbarik/alaska [Access on 8th August, 2016]. http://nationalmap.gov/small_scale/atlas-ftp-globalmap.html?openChapters=chpbound#chpbound[Accesse d on 8th August, 2016] [23]. Zhu, Haijun, and Chaowei Phil Yang, "Data Compression for Network GIS," Encyclopedia of GIS, Springer US, pp. 209-213, 2008. [24]. Chen, F., & Ren, H., “Comparison of vector data compression algorithms in mobile GIS,” 3rd IEEE International Conference on Computer Science and Information Technology (ICCSIT), 2010. [25]. Ji, Huifeng, and Yihe Wang, "The Research on the Compression Algorithms for Vector Data, " IEEE International Conference on Multimedia Technology (ICMT), 2010. [26]. http://nationalmap.gov/small_scale/atlas-ftp-globalmap.html?openChapters=chpbound%2Cchpwater#chpwater [Accessed on 8th August, 2016]