Cloud Computing - Information Technology and Productivity
|
|
Cloud Computing - Information Technology and Productivity |
| Cloud Computing Traffic - Projecting Growth | Major Players in Cloud Computing | Strategies | |
| Cloud Computing Services | Cloud Computing - Emerging Trends | Web Browser Viewing - MS Edge |
|
Information technology has entered an age of compound and hybrid technologies. Cloud computing is a conglomeration of computer technologies which provide access to computing, data storage, and communications:
| Integrated Systems of Hardware and Software |
Open Standards, Performance, and Shared Information |
Open Source versus Commercial Software Computing Models |
|
Increasingly computer processing and storage is being rented from cloud computing service providers. Cloud companies receive approximately 10 percent of all spending on business computing; and growth is occurring at 30 percent per annum. There are more than 30 billion internconnected objects on the Internet.1 In the first quarter of 2018, spending on public-cloud infrastructure increased by 32% from the prior year period. To meet rising demand for public cloud services, providers are spending on hardware, software, and other infrastructure components for data center infrastructure: servers, storage, networking, operating systems, virtualization software, network security, and management software. Global spending on public cloud services and infrastructure is projected to increase to $160 billion. Banking, professional services, and manufacturing are leading the transition to the cloud. The United States is the largest market for the public cloud; it accounts for approximately 60% of global spending. 1 Organizational enterprise and new entrants are evaluating system and application software in relation to their business model, the needs of clients, and the economics associated with going directly to the next generation of commercial and open source software technology. Within this operational reality, cloud computing is evolving from concept to a practical implementation. It is shifting the emphasis from locally managed server-client installations and information technology related services to externally located web-accessible computing centers consisting of thousands of servers. However, there is no single uniform precise definition of cloud computing and there are different types of clouds. The three standard models of cloud computing platform service are: SaaS: Software as a Service, IaaS: Infrastructure as a Service, PaaS: Platform as a Service. There also is a IaaS/PaaS; a hybrid service in which applications can be developed on a cloud-based operating system and then hosted in a consolidated software environment: operating system, database, and application. When moving from traditional hardware to the cloud, the implementation will be to IaaS, PaaS, or customer hardware options. Important considerations will be the complexity of the application and the amount of control required. Simple applications can be deployed on PaaS and more complex applications on IaaS. There will be a requirement to track, monitor, and visualize the necessary aspects of application. The right kind of management framework will be required in order to abstract away the low-level management into higher-level constructs, such as managing a single stack or services as one unit. Cloud APIs: application programming interfaces are a software technology that allows for communication between applications or infrastructures in the cloud. In 2009, Zend, Microsoft, IBM, Rackspace, Nirvanix, and GoGrid collaborated on the open-source Simple Cloud API to interact with different cloud providers. This allowed PHP developers to access services on many clouds. The early adapters were for file-storage services, document-storage services - MS Windows Azure Table and Amazon Simple DB support and simple queue services- Amazon SQS and MS Azure queues. However, what has transpired since then is that a leading vendor emerges in a cloud computing category and ends up developing a successful standard. This bypasses the creation of a study group and a standards committee and the long process of releasing an international standard. Many current standards started as proprietary cloud APIs and then evolve into industry-wide standards established by the market leader. Cloud application programming interfaces are driving the increase in application development and creating cloud interoperability. Effective APIs enable developers to reconfigure applications and integrate components without having to change or write new code. Cloud APIs commonly expose their features using REST: Representational State Transfer and SOAP: Simple Object Access Protocol. |
||||
The growth of cloud services is being driven by:
There are research firms forecasting that the global cloud computing market will grow from $41 billion in 2011 to $241 billion in 2020. Of that amount, the projections by 2020 for the North American market is $47 billion and the European Union (28 member countries) market is $16.1 billion.2 The Cisco Systems Global Cloud index published in November 2011 has proven to be accurate.3 Both global cloud and data center traffic overall experienced a significant compound annual growth rate between 2010 and 2015. The data indicates that currently in the range of 50 percent of all workloads are being processed in the cloud and in 2017 - 2018 there will continue to be an increase the cloud based data center traffic. With 2010 as a baseline, Cisco Systems reported that global data traffic was in the range of 1.1 zettabyte. In 2015, the average data traffic was estimated to be 4.8 zettabytes. According to executives at Cisco Systems, most countries currently can support some level of cloud services based on average download and upload speeds and the time it takes for information to pass through the network for business and consumer connections. However, not all regions within a country currently are able to support the high-end advanced cloud apps such as high definition video conferencing. It will take time and investment in global network expansion to handle advanced cloud services. Public clouds will have to address the management of security and regulatory compliance. There will be multi-cloud formation and then mega-clouds. Mega-clouds will weave together multiple private and public clouds to work as a coherent, holistic system. Mega-clouds will require innovations in networking to move data between clouds, storage to place data in the right cloud, processing for efficient workload computing, and orchestration to link networking, storage and compute together across clouds. The assumption is that it will be needed to incorporate artificial intelligence and machine learning to bring automation and insight to business models and government planning.
The major players have been developing cloud computing services based upon strategic interest and its business model.
Amazon Web Services and Microsoft's Azure are the leading public cloud providers. The forecast is for the market leaders, AWS and Azure, to exceed the expected market growth. There also is brand recognition and positive purchase intent for Google Cloud and IBM Cloud services. According to Goldman Sachs, by the end of 2019, enterprise workloads running on public cloud services offered by the three leading cloud services - Amazon, Microsoft, and Google will have increased from 18% to 34%. The open source OpenStack Foundation and Project provides a framework with a set of APIs and tools into which products and technologies can be integrated to create private, public, and hybrid cloud services. OpenStack started with the Nova Compute and Swift Storage projects. Within Nova multiple forms of hypervisors and virtualization can be deployed: open source KVM, VMware ESX, and, Microsoft's Hyper-V are supported. The OpenStack Integrated Release has been expanded to include multiple projects. The OpenStack Kilo release, the integrated projects include: Nova compute, Swift object storage, Cinder block storage, Keystone identity, Horizon dashboard, Glance image, Neutron networking, Trove database, Sahara Big Data, Heat orchestration, Ceilometer monitoring, and Ironic Bare Metal projects. In terms of support and deployment, the OpenStack cloud has emerged as the leading private cloud platform. Dell-EMC Pivotal Cloud Foundry is a set of software development blueprints and tools to build and deploy new cloud-based applications. Promoting iterative software development, Cloud Foundry is based on and also available as open source software code. An enterprise can run software applications on internal VMware infrastructure or beyond its firewall on AWS: Amazon Web Services, Alphabet-Google, or Microsoft Azure cloud platforms without vendor lockin. Key issues and areas of contention which have emerged are open standards, shared information, and performance. Google uses open standards, but also keeps many things proprietary. Engineers in the company have acknowledged that Google has developed specialized semiconductors for some of its in-house operations. According to the online New York Times, it has selectively not been patenting products, as part of a competitive strategy and better for service its customers.4 Facebook has published specifications of almost everything in its data centers, from printed circuit boards to building cooling systems. The idea is that others can use parts of this technology themselves, or possibly contribute to its improvement. The Oracle Corporation strategy is based upon the assumption that engineering hardware and software together will provide better performance, ease of use, and lower cost. The Oracle cloud has been developed based upon open source standards with data services, Java services, and security services built in. Large corporate databases can be moved from a company's machine into the Oracle cloud or created in the Oracle cloud and moved back and forth into the clouds run by Amazon, IBM or other computing service providers running in Java. The Oracle cloud places in it direct opposition with Salesforce. The emergence of cloud computing has not been a unifying force in information technology. Rather than making the consumer and business technology increasingly similar, the computer world is splitting into two separate camps: consumer and enterprise. In the consumer camp are Apple, Google, and most of the Asian hardware makers. The enterprise camp will service large companies and government agencies; the major players will be IBM, Dell, Cisco, and Oracle. Microsoft and Hewlett-Packard will straddle both markets. The 250 leading cloud organizations have established the CSCC: Cloud Standards Customer Council in order to collaborate and address the challenges associated with implementing cloud computing. New companies are emerging which provide expertise in analyzing, developing, and integrating applications on the cloud and mobile devices. There can be thousands of different applications accessing the cloud of a data center's computer servers and storage devices. It no longer is straightforward to know where software initiates from and terminates. A typical application will be executed over multiple servers and pass through several layers of the software stack. Neither will the infrastructure processing the application be static. There are multiple layers of software which handle authentication, middleware to other applications and systems, retrieving and placing data, and communicating and responding with globally located systems. Realtime products offer a way to monitor the performance of applications in the cloud; this can result in a significant improvement from the daily or random checks being performed at data centers. Equally as important, these services provide predictive descriptions where slowdowns can occur and how to achieve operational savings. Coding and development on public clouds will have applications being integrated from geographically dispersed data centers. Sound design and intelligent implementation will be required to optimize performance associated with the delivery and processing of the application and controlling costs.
Applying the ideas and principles of cloud computing to information technology operations can offer organizations agility and cost savings. The huge explosion of data is driving a movement to design systems around workloads in order to deliver the required computation and service and become more energy efficient. For many enterprises, its information technology infrastructure is under stress and the data flood is accelerating. Storage, delivery, and processing of data now can be selected from or be a combination of cloud computing, traditional outsourcing, and in-house personnel. The hardware and software languages, development platforms, and scalable databases being used in the cloud reflect competitive and evolutionary forces of commercial software providers, open source software alternatives, intellectual property laws, anti-trust regulations, and international governing bodies. However, despite the benefits a cloud environment promises, there are issues that will need to be addressed regarding the security and privacy of enterprise applications. Unlike the investment in commercial information technology infrastructure, which in many situations has been developed and implemented over a half century as part of the modern era of data processing, these cloud technologies may not always be categorically or immediately better than a web-enabling and retrofitting of existing applications and system software. SYS-ED has been aggregating information and analyzing trends:
|
||||
|
Footnote 1: Assumptions and numerical projections were provided by: Bernstein Research, reported by the New York Times online - September, 2016. McKinsey Figures - Bloomberg Business Week - November 2016. Forrester Research, International Data Corporation, Synergy Research Group statistics reported in the CIO Journal - Wall Street Journal by Angus Loten, June 18, 2018.
Footnote 2:
Footnote 3:
Online New York Times - Quentin Hardy - multiple reports.
Sources: Wall Street Journal reporting by Dan Gallagher. |
