60.4 Open Issues and New Trends
There is a large gap between current technologies in Big Data and companies from a great diversity of business domains that would require incorporating these tools into their value chain. It is mainly due to the difficulties entailed in facing the challenges presented by having multi‐domain, business‐wide solutions, or the need to hire experts to build and manage them outside the core business. Besides, this is in accordance to investment costs and deployment time constraints. Even though workflow technology pretends to become an interesting alternative to allow companies to model and execute their own business‐specific data analysis, there are still some open issues that require further discussion.
Firstly, the wide adoption of custom workflow‐based Big Data solutions requires the adaptation of WfMS to domain‐specific data analysis processes. As a way to mitigate development costs, some WfMS already provide a number of different configurable components to be reused in a variety of business domains. However, this configuration process is still limited to similar application domains with similar requirements, implying a lot of development effort and technological skills. More abstract, intuitive and ready‐to‐use workflow tools on their own specific domains rather than generic and complex WfMS are demanded by business experts.
Additionally, notice that running Big Data workflows is an extremely demanding task in terms of resource requirements. The infrastructure required to efficiently execute data‐intensive procedures demands new approaches to lower operating and maintenance costs. Cloud computing seems a suitable solution here for executing workflows according to an on‐demand model, avoiding infrastructure facilities, management costs and technicians. Nevertheless, this approach involves considering that WfMS should provide cost‐aware mechanisms to schedule the optimal workflow execution according to a well‐planned, but restricted, use of resources like computation power or data transfer rates.
Designing a workflow is a manual, time‐consuming and error‐prone task. Notice that in many contexts, it is hardly possible to represent and configure a correct domain‐specific data processing and to decide on appropriate analytical methods without deep knowledge about these methods. As previously mentioned, workflows should be represented using an user‐friendly notation, meaning that business experts would be able to easily achieve the necessary skills to model their data requirements. Even further, such a representation should be formally defined in terms of a well‐defined language at a high level of abstraction, offering new opportunities for automating the development of data analysis processes and increasing the flexibility, portability, easiness to learn and interoperability of the resulting workflows. Additionally, combining well‐defined workflows with semantic technologies might enrich their specification with domain‐specific properties, enabling the inline validation of workflows or the discovery of the aforementioned ready‐to‐use components.
New trends are focusing on separately addressing these issues. In this context, WorkGenesis13 is a novel framework to simplify the creation of new custom WfMS, which enables the definition and execution of business‐specific data analysis processes. WorkGenesis uses MDE (model‐driven engineering) and highly decoupled components to automatically generate custom WfMS by reusing pieces of knowledge from different workflow definitions, as well as software components and the declaration of previous data types and processes. It provides a set of domain‐agnostic configurable software elements that can be selected, adapted and reused within the specific business domains. The resulting WfMS provides ready‐to‐use actions and resources, fully adapted to each business expert’s particular interests, who only needs to focus on the specification instead of the technical development.
WorkGenesis is divided into three layers: (1) a meta‐tool for the automatic generation of domain‐specific WfMS from domain‐independent components and resources; (2) a customizable workbench to draw and assemble workflows including those components, previously configured by the meta‐tool; and (3) a flexible, portable and extensible workflow engine, responsible for deploying the data analysis processes and resources into different computational models, e. g., from a local parallel execution to a distributed grid‐based platform like Apache Hadoop. As a result, WorkGenesis facilitates the creation of solutions for specific business domains by reusing knowledge assets (e. g. predefined nested workflows or imported processes), saving development costs and reducing the time to market.
60.5 Concluding Remarks
Big Data is causing a paradigm shift in industry, where increasingly often large amounts of data are produced in or retrieved from heterogeneous data sources. In fact, it is expected that 75% of companies will invest in these technologies within the next two years. Actually, Big Data solutions have already being successfully applied to domains like health, manufacturing, governance, education, banking and others. Nevertheless, the excessive product offer, associated cost and complexity of current technologies, platforms and tools may hamper its business‐wide adoption, making it dependent on the investment options of each specific sector.
This chapter has discussed about the difficulties found to generalize the use of Big Data solutions to business experts. However, workflows have already proved to be applicable to and cost‐efficient in diverse domains as a way to bring these experts closer to how they usually specify and represent their own application domains. Workflows allow defining the specific sequence of actions to be performed during the data analysis process. Apart from providing a visual notation, WfMS can manage the efficient execution of each workflow, transparently to the business expert. However, there is still a number of open issues to address, such as the lack of standard solutions for data‐intensive applications like Big Data or the need of highly specialized teams for undertaking new developments, among others. WorkGenesis has been introduced as a novel solution to overcome these shortcomings by automatically generating custom WfMS and reducing the time to market.
References
1.
R. van der Meulen and V. Woods, “Gartner survey shows more than 75 percent of companies are investing or planning to invest in Big Data in the next two years,” Gartner, 2015. [Online]. Available: http://www.gartner.com/newsroom/id/3130817.
2.
D. Simchi-Levi, J. Gadewadikar, B. McCarthy and L. LaFiandra, “Winning with analytics,” Accenture, 2015.
3.
M. Turck, “Is Big Data still a thing? (The 2016 Big Data landscape),” FirstMark Capital, 2016. [Online]. Available: http://mattturck.com/2016/02/01/big-data-landscape.
4.
D. Loshin, “Achieving organizational alignment for Big Data analytics,” in Big Data Analytics, Morgan Kaufmann, 2013, pp. 21–28.Crossref
5.
Workflow Management Coalition, “Terminology & Glossary,” 1999.
6.
J. Yu and R. Buyya, “A taxonomy of workflow management systems for grid computing,” Journal of Grid Computing, vol. 3, no. 3, pp. 171–200, 2006.
7.
R. Frye and M. McKenney, Information granularity, big data, and computational intelligence, Springer, 2015.
8.
T. White, Hadoop: The definitive guide, O’Reilly Media, 2015.
9.
IBM Software, “Data-driven healthcare organizations use big data analytics for big gains,” 2013.
10.
B. Kayyali, D. Knott and S. Van Kuiken, “The ‘big data’ revolution in healthcare: Accelerating value and innovation,” McKinsey & Company, 2013.
11.
D. Adamson, “Big Data in healthcare made simple: Where it stands today and where it’s going,” [Online]. Available: https://www.healthcatalyst.com/big-data-in-healthcare-made-simple. [Accessed 10 08 2016].
12.
W. Liu, Q. Li, Y. Cai, Y. Li and X. Li, “A prototype of healthcare b
ig data processing system based on Spark,” in 8th International Conference on Biomedical Engineering and Informatics, 2015.Crossref
13.
S. Rallapalli, R. R. Gondkar and U. P. K. Ketavarapu, “Impact of processing and analyzing healthcare Big Data on cloud computing environment by implementing Hadoop cluster,” Procedia Computer Science, vol. 85, pp. 16–22, 2016.Crossref
14.
L. Skiftenes Flak, W. Dertz, A. Jansen, J. Krogstie, I. Spjelkavik and S. Ølnes, “What is the value of eGovernment – and how can we actually realize it?,” Transforming Government: People, Process and Policy, vol. 3, no. 3, pp. 220–226, 2009.Crossref
15.
E. Kalampokis, E. Tambouris and K. Tarabanis, “A classification scheme for open government data: Towards linking decentralised data,” International Journal of Web Engineering and Technology, vol. 6, no. 3, pp. 266–285, 2011.Crossref
16.
The White House, “Transparency and Open Government. Memorandum for the heads of executive departments and agencies,” 2009. [Online]. Available: https://www.whitehouse.gov/sites/default/files/omb/assets/memoranda_fy2009/m09-12.pdf.
17.
V. Vijayakumar, V. Neelanarayanan, J. Archenaa and E. A. Mary Anita, “Big Data, cloud and computing challenges A survey of Big Data analytics in healthcare and government,” Procedia Computer Science, vol. 50, pp. 408–413, 2015.Crossref
18.
G.-H. Kim, S. Trimi and J.-H. Chung, “Big-data applications in the government sector,” Communications of the ACM, vol. 57, no. 3, pp. 78–85, 2014.Crossref
19.
B. Esmaeilian, S. Behdad and B. Wang, “The evolution and future of manufacturing: A review,” Journal of Manufacturing Systems, vol. 39, pp. 79–100, 2016.Crossref
20.
Oracle Enterprise Architecture, “Improving manufacturing performance with Big Data. Architect’s guide and reference architecture introduction,” 2015.
21.
M. Gaitho, “How applications of Big Data drive industries,” 2015. [Online]. Available: http://www.simplilearn.com/big-data-applications-in-industries-article.
22.
C.-S. Neumann, “Big data versus big congestion: Using information to improve transport,” McKinsey & Company, 2015.
23.
B. Marr, “Big Data: The winning formula in sports,” Forbes, 2015.
24.
Y. Zhang and Y. Zhao, “Astronomy in the Big Data Era,” Data Science Journal, vol. 14, 2015.
25.
Deloitte, “Opportunities in telecom sector: Arising from Big Data,” 2015.
26.
M. R. Berthold, N. Cebron, F. Dill, T. R. Gabriel, T. Kötter, T. Meinl, P. Ohl, C. Sieb, K. Thiel and B. Wiswedel, “KNIME: The Konstanz Information Miner,” in Studies in Classification, Data Analysis, and Knowledge Organization, Springer, 2007.
27.
M. Hofmann and R. Klinkenberg, RapidMiner: Data mining use cases and business analytics applications, Chapman & Hall/CRC, 2013.
28.
K. Wolstencroft, R. Haines, D. Fellows, A. Williams, D. Withers, S. Owen, S. Soiland-Reyes, I. Dunlop, A. Nenadic, P. Fisher, J. Bhagat, K. Belhajjame and F. Bacall, “The Taverna workflow suite: Designing and executing workflows of web services on the desktop, web or in the cloud,” Nucleic Acids Research, vol. 41, no. W1, pp. W557–W561, 2013.Crossref
29.
KNIME, “Outlier detection in medical claims,” [Online]. Available: https://www.knime.org/knime-applications/outlier-detection-in-medical-claims. [Accessed 15 08 2016].
Further Reading
30.
D. C. Schmidt, “Guest editor’s introduction: Model-driven engineering,” Computer, vol. 39, pp. 25–31, 2006.Crossref
Footnotes
1Apache Pig. https://pig.apache.org.
2Apache Spark. http://spark.apache.org.
3Apache Hive. https://hive.apache.org.
4Apache HBase. https://hbase.apache.org.
5Cloudera Impala. https://cloudera.com/products/apache-hadoop/impala.html.
6Microsoft Azure Machine Learning. https://azure.microsoft.com/en-us/services/machine-learning.
7Amazon Web Services. https://aws.amazon.com.
8IBM Watson. http://www.ibm.com/watson.
9IBM Watson Health. http://www.ibm.com/watson/health.
10IBM InfoSphere. https://www-01.ibm.com/software/data/infosphere.
11IBM BigData. https://www-01.ibm.com/software/data/bigdata.
12Xplenty. https://www.xplenty.com.
13WorkGenesis. http://www.workgenesis.com.
Part XIV
Cloud Technologies
© Springer-Verlag GmbH Germany 2018
Claudia Linnhoff-Popien, Ralf Schneider and Michael Zaddach (eds.)Digital Marketplaces Unleashedhttps://doi.org/10.1007/978-3-662-49275-8_61
61. Preface: the “Cloud Way” to Digital Transformation and New Business Models
Sabine Bendiek1
(1)Microsoft Deutschland GmbH, München Schwabing, Germany
Sabine Bendiek
Email: [email protected]
Tomorrow, traffic will be safer, more sustainable and more efficient. Autonomous vehicles will move around without direct driver input to transport people and goods, on demand, from door to door using the most efficient routes. They will interact with other transport systems, offering seamless end‐to‐end journey connectivity and producing convenient and affordable mobility to everybody.
Tomorrow, healthcare will be more reliable, more individual and more preventive. The access to a massive amount of digitized data combined with analytics and artificial intelligence will enable researchers as well as physicians to develop more specific measures and more personalized treatments – enhancing medical care on a large scale.
Tomorrow, manufacturing will be smarter, more agile and more resilient. Connectivity and the internet of things will allow companies to react faster and more directly to customers’ needs, producing and delivering better products at lower costs – creating better value for businesses and consumers all over the world.
Tomorrow, work will be more flexible, more mobile and more productive. New technologies will enable virtual teams to collaborate without boundaries and to seamlessly share knowledge and ideas – enhancing freedom of thinking and creativity.
Similar changes will soon occur in many other areas and profoundly affect our lives. Enabling that transformation are intelligent systems that help us gain insight and take action from large amounts of data – based on the power of cloud computing (see Fig. 61.1).
Fig. 61.1Identify the right opportunities to drive your digital transformation
61.1 Democratizing Companies and Markets
Cloud computing enables ubiquitous, on‐demand access to computing power – databases, servers, storage networking, software, analytics and more – at low costs and high speed over the internet. It relies on the sharing of resources to achieve coherence and economy of scale over a network. It allows companies and organizations to get their applications running faster, to adjust resources and thus becoming more flexible, and to focus on their core business instead of infrastructure. Cloud computing offers high computing performance, scalability and availability. In the cloud even vast amounts of computing resources can be provisioned within minutes. The cloud enables small companies to benefit of similar technological means as large multinationals, providing them with new competitive advantages and thus democratizing markets. Cloud computing facilitates the access to data and services by offering full device and location independence, as users can connect anytime and anywhere and work on the same data simultaneously. In the cloud, knowledge becomes transparent and accessible to everybody – it can no longer be exploited as an instrume
nt of power or misused to mark hierarchical differences – thus democratizing companies from the inside.
61.2 Creating Disruptive Business Models
Cloud computing marks a big shift from the way we used to think about IT – and while some organizations only just start to understand to what extent it will change our world, others have already fully realized the power of the cloud. They have set up new business models that deeply transform consumer behavior and expectations within a very short time. These new businesses based on the power of cloud computing typically act like platforms bringing together producers and consumers in high‐value exchanges. Their chief assets are information and interactions – perfectly matched in the cloud they create amazing value: Facebook, the biggest news‐machine in the world, does not produce any content.
Uber the largest taxi‐service, owns no cars.
AirBnB, the prevalent lodging‐provider, operates no hotels.
Digital Marketplaces Unleashed Page 93