Industry Trends and the Future of Cloud Computing - The Enterprise Cloud: Best Practices for Transforming Legacy IT (2015)

The Enterprise Cloud: Best Practices for Transforming Legacy IT (2015)

Chapter 9. Industry Trends and the Future of Cloud Computing

Key topics in this chapter:

§ Analysis of industry trends

§ Evolution from IaaS to PaaS applications

§ Convergence of IaaS and SaaS providers

§ Trends leading from private to hybrid clouds

§ The future of cloud computing:

§ Hybrid clouds and cloud brokering

§ Application transformation: fully multithreaded, multiprovider, dynamically scalable applications

§ Self-service administration: consolidated application control panels

§ Software-defined datacenters

§ Big Data and analytics

§ The Internet of Things

The cloud computing industry has matured quickly in its initial years, and certainly since the term “cloud” was first coined. As the pace of computers and information technology in general continues to accelerate. The cloud industry has clearly gone through its infancy and is in what I would call the adolescent stage. This book has focused on providing lessons learned during these early years, and now we look to the next generation of cloud computing.

In this chapter, I first discuss some important industry trends, some of which are lessons learned by cloud providers and early-adopter customers. Then I will discuss the technologies, trends, and forecasts for the next-generation clouds. By understanding these trends and technologies, you and your organization can improve your cloud transition, better comprehend the current cloud industry, and build a more flexible future-ready enterprise cloud.

An Analysis of Industry Trends

As I’ve discussed throughout this book, cloud computing can be considered as the modern incarnation of legacy datacenters embodying a new style of IT operations, procurement, and management. The concepts of virtualization, rapid provisioning, pay-as-you-go, and elasticity are no longer new concepts; they are now common benefits and techniques employed by public cloud providers and operators of private clouds. To determine where the industry is headed and the future of cloud computing, it is important to evaluate the latest trends.

Public Cloud Providers

Most of the major public cloud providers such as Amazon AWS originally focused on providing Infrastructure as a Service (IaaS), virtual machines (VMs), and web hosting and content delivery network (CDN) services. Then, these providers began to add Platform as a Service (PaaS) and Software as a Service (SaaS) offerings. IaaS is now so commonplace that every public cloud provider is expected to have this capability, and price is one of the few differentiators between competitors.

SaaS Providers

A mere five years ago in 2010, cloud providers with a SaaS focus were called application service providers (ASPs). These cloud service providers and software-management platforms are quickly expanding into PaaS and IaaS applications, seemingly taking the opposite approach used by Amazon, which started out with IaaS and then moved to PaaS. These SaaS providers have the advantage of mature and feature-rich self-service application control panels, but they do not have the size, scalability, and investment funding that the big IaaS public cloud providers enjoy.

The Convergence of the IaaS and SaaS Industries

The primarily larger IaaS-focused providers and the often-smaller SaaS providers surprisingly have not merged or had any significant rounds of acquisitions yet. Providers such as Microsoft have a significant internal portfolio of both SaaS, IaaS, and PaaS technologies that it is transforming into cloud offerings. Using Microsoft again as the example, it started with the SaaS-based Office 365, and then later launched its Azure cloud service, which includes IaaS and PaaS offerings based on its Hyper-V, SQL, and other platforms. Microsoft has yet to combine some of its SaaS offerings (such as Office 365) with its Azure public cloud in an integrated unified interface. Google also began with mostly SaaS applications (such as Google Apps). More recently, it has added IaaS offerings via the launch of Google Compute, but the Google Apps SaaS offering is still a separate product. Throughout the industry, you can expect to see more SaaS and PaaS products offered by traditional IaaS providers, and numerous acquisitions of small SaaS providers by the industry giants — particularly because competitive pricing for basic IaaS applications continues to drop, whereas PaaS and SaaS is projected to be profitable in the long term.

Systems Integrators

IT systems integrators have mostly skipped the public cloud business and focused more on customizable and secure private clouds — often built as dedicated enterprise clouds for mostly larger customers and public sector government agencies. Hewlett-Packard has notably continued its private cloud focus for customers while also entering the public cloud market, stating publicly that the combination of private and public cloud services — hybrid cloud — is where the industry is headed and customers want to go. Expect systems integrators to continue acquiring cloud management software companies to differentiate themselves. These platforms make up the basis of automation, orchestration, billing, and customer portals deployed by the systems integrators to their customers. In other words, systems integrators will continue to cloud-enable individual enterprise customers or build private/hybrid clouds, but most of them will not become cloud providers themselves.

Private to Hybrid Cloud

Although most cloud consumers want all of the benefits of a public cloud (such as elasticity, low cost, little or no commitment, pay-as-you-go, etc.), the ability to customize a cloud and implement specialized security controls has made private cloud more popular.

The industry-wide trend points to private cloud customers using one or more Anything as a Service (XaaS) products from public cloud providers, as well; so the hybrid cloud model is looking to be the most common form of cloud in the coming years. The next generation of customer clouds will be a primarily hybrid cloud model whereby customers own and operate a private cloud — having modernized legacy datacenters — and use the private cloud management platform to broker out selected services to third-party public cloud IaaS, PaaS, or SaaS providers. These hybrid clouds will become the norm in the industry, so much so, in fact, that even pure public cloud providers will adapt and offer more hybrid and brokering-type cloud integration to other cloud providers and partners through application programming interfaces (APIs) and online marketplaces.

Key Take-Away

The next generation of enterprise clouds will be primarily hybrids whereby customers own and operate a private cloud — usually by first modernizing legacy datacenters — and use the cloud management platform to broker selected services to third-party public cloud IaaS, PaaS, or SaaS providers.

Small Business

Small businesses — typically those that have fewer than 100 employees — are more likely to consume public cloud services over the Internet. These companies have little or no IT department but still want to benefit from IaaS and SaaS providers to reduce the costs and burdens of in-house computing. Many small-business startups have no internal IT personnel and must completely outsource their needs to public cloud providers. Private clouds are often out of the question for small businesses due to the cost, complexity, and lack of adequate IT skills (or willingness to establish an IT department).

Purchasing Cloud Services

Small and medium-sized organizations often utilize credit cards for their cloud service purchases. The largest customers, and particularly public sector government customers, tend to use purchase orders submitted to the cloud provider with invoices sent by the cloud provider back to the customer at the end of each billing period. A clearly observable trend is that credit card–based purchases are often associated with rogue or shadow IT — when individuals or teams within an organization purchase a cloud offering on their own, often to reduce the time and complexity of acquiring new IT assets through internal company processes.

Cloud Deployment Models

Government customers tend to deploy mostly private clouds for the majority of their services, whereas public cloud is often used for limited use cases such as development and testing or for relatively static public websites. Private and hybrid clouds often require more capital investment or commitment on behalf of the customer organization but gain increased customization capabilities. Community clouds, with multiple cooperative peer organizations sharing a common cloud ecosystem, are not very popular so far, because peer organizations often have difficulty agreeing among themselves on governance, funding, roles and responsibilities, as well as the significant concern that any individual peer’s future funding, mission, or leadership changes could disrupt the community.

Key Take-Away

Community clouds have inherent organizational, financial, political, and long-term viability complications. A community cloud is a lot like a marriage: a great idea at the beginning, but will it hold up over time if and when peer agencies change leadership, mission, or funding?

The Future of Cloud Computing

The theme throughout this book has been to analyze lessons learned in the first generation of the cloud and to provide best practices with a focus on technology and business for the next generation. Based on lessons learned, industry trends, and customer-demand signals, the next-generation clouds will focus on a new style of delivering IT rather than the technology itself.

Key Take-Away

The next-generation clouds will focus on a new style of delivering IT rather than the technology itself. The business outcomes and focus on customer demands will clearly be the nexus point, whereas the underlying technologies will still be important but no longer the business justification.

Although the first generation of clouds focused on basic concepts such as pay-as-you-go, elasticity, ubiquitous access, and ease of use, next-generation clouds will expand this list of characteristics for all clouds — essentially expanding the baseline standards and expectations for future clouds.

The Hybrid Cloud

Public and private cloud deployment models are currently the most popular choices for customers. Based on lessons learned and customer-demand signals, private clouds are favored for their customization and parallel migration paths for legacy datacenters. Soon after organizations begin their transition to the cloud, clarified requirements and legacy application needs often cause the organization to consider using a combination of on-premises and private cloud services along with external public clouds for selective XaaS needs. This combination is, by definition, a hybrid cloud — the ability to use selected public cloud XaaS offerings through the same private cloud management portal.

The trend continues toward private cloud as the baseline service for many organizations with the ability to burst or hybrid provision to selected public cloud services on demand. This hybrid cloud requires new focus on automated and dynamic provisioning of cloud services across multiple cloud providers and legacy datacenters. The ability to provision in this next generation of cloud requires new technology and tools to perform service aggregation and arbitration of XaaS, which is the basic technology behind hybrid and cloud service brokering.

Key Take-Away

Hybrid clouds will become so common that the term “hybrid cloud” might fade away, because it will be the norm for modern IT delivery.

Most customers — the larger organizations and government entities in particular — want all of the characteristics and benefits of public cloud offerings. As detailed requirements are evaluated, these large organizations quite often determine that they would prefer a private cloud. The primary benefits of private clouds include customizations to the procurements, security, operations, reporting, and governance processes. Keep in mind that public cloud services provide 90% of their services as standardized products, with limited customization. Only private cloud deployments have the ability to customize some or all aspects of the service to meet customer requirements. Private clouds also best facilitate the modernization and migration to cloud transformation for datacenters on a slower transition schedule. So, each organization is left with the decision to sacrifice their customization needs and go with a public cloud or embrace their uniqueness and move to a private cloud.

Key Take-Away

Many customers say they want the features and benefits of a public cloud. After discussing detailed customization, security, and other requirements, though, it often becomes clear that a private cloud model is really the more appropriate one.

As stated earlier, public cloud providers are also creating hybrid services whereby the public cloud integrates back to legacy enterprise datacenters — essentially the opposite of hosting your own private cloud management platform and extending it to public cloud XaaS. I expect to see a massive uptake in the sophistication and quantity of these public-to-enterprise integration tools to further blur the line between on-premises enterprise datacenters and public cloud services.

Bridging Public Cloud to Enterprise Datacenters

A new and powerful capability being introduced by some public cloud providers is the ability to reach back into a customer’s enterprise datacenter. This is done by federating the authentication system (such as Active Directory) and connecting the public cloud provider’s network with the customer’s network. These network connections can be made through the Internet by using a virtual private network (VPN) or a dedicated wide area network (WAN) communications circuit. After the public cloud and enterprise cloud are connected, consumers of the cloud can then use any combination of pubic and legacy applications, log on using the corporate directory password, and synchronize data across the cloud and the legacy datacenter.

This approach — in which a public cloud provider enables a hybrid cloud — is very new in the industry but shows a significant potential to provide the following hybrid services:

§ Backup of enterprise on-premises servers to the cloud

§ Federated identity management and authentication between the cloud and enterprise directory

§ Cloud-based mobile device management

§ Remote application publishing from the cloud to any end-user desktop or mobile device

§ Synchronization of cloud and enterprise data that can be accessed from any device and any location via the cloud

§ Redundancy and failover of enterprise servers to cloud-hosted VMs

Software-Defined Datacenters

The software-defined datacenter (SDDC) is an approach to configuring IT infrastructure through virtualization of servers, networks, and storage. All of these infrastructure services are pooled and the individual hardware components abstracted behind virtualization software. In an SDDC, the server, storage, and network resources are managed by a cloud management platform that performs automated configuration, provisioning, and allocation of resources.

The three core technologies in an SDDC are virtualization of servers, storage, and networking. Technically, many other components of the infrastructure can be virtualized and automated, such as security, operations, and even components of the physical datacenter facility. As is covered inChapter 3, the virtualization of physical servers is already commonplace for modern datacenters and cloud environments. The focus in the future will be on virtualizing of storage and networking.

Network virtualization involves pooling all network switches, routers, firewalls (along with IP addressing), DNS, and other network-related services. The cloud automation software automatically configures network segments, routing paths, and firewall rules on demand as customers order their cloud services. The physical network cabling does not run from a server network card directly to a central switch, as in a traditional datacenter; instead, multiple server blades are concentrated into a high-density server chassis, with the chassis or server cabinet having multiple shared high-speed connections to the network. Then, one or more virtual connections are made, through software, to create the virtual network path between the server and the rest of the network, to a specific subnetwork, or other infrastructure components. So, the physical network is configured in an any-to-any configuration so that the software is used to actually connect the networks paths, creating and changing these paths through the automation software to keep things manageable. You can create, change, or remove these software-defined paths dynamically through the network via the cloud automation software as needed. The same is true for storage virtualization where server farms are not necessarily directly connected to a SAN; instead, the automation software creates a virtual path between the server(s) and the abstracted pooled storage system(s).

The benefits to these software-defined techniques are numerous. The primary reason is to enable a large quantity of servers, network, and storage infrastructure to be physically deployed and ready for automated configuration just in time upon a cloud service being ordered. The automated provisioning systems actually assign processors, memory, storage, and network configurations to each server or virtual server. One server might be configured to host IaaS virtual machines (VMs), whereas another might be configured to run an enterprise SaaS application — all depending on customer demand and what is ordered. In this example, servers are no longer preconfigured with application software but are instead left in a generic “blank” state, ready for instructions on what hypervisors, operating systems (OSs), and applications to load after a customer places an order.

Key Take-Away

The concept of manually configuring every server with network, storage, OS, and applications is now antiquated. Automation tools now create virtual servers, network, and storage paths on demand, as customers order services. Essentially all things are now dynamic and software defined rather than static, hardcoded, or manually configured.

Some other benefits to “software-defined everything” is that you can swap out servers easily and the replacement server automatically inherits the characteristics and configuration of the original. And as more capacity is needed for the cloud infrastructure, you can add more unconfigured servers, network, and storage devices into the datacenter with the automated and software-defined configurations automatically applied as customer orders arrive and begin to utilize the newly installed capacity.

Finally, the SDDC give you the ability to span services across multiple datacenters for extra capacity and redundancy. The applications and end users of the cloud have no knowledge that the cloud services, storage, and networking spans one or more datacenters or when a failover to a disaster recovery facility has occurred. This technique also makes a cloud service portable — you can move it from one cloud provider to another with the software-defined network redirecting all users to the new provider or datacenter with little or no service interruption.

Taking the software-defined concept to the datacenter level, you can dynamically control traditional physical systems such as HVAC, fans, and lighting via software. Imagine datacenter facilities automatically adapting to increases in customer demand by adjusting environmental settings and fans to maintain proper, balanced temperatures. Or perhaps turning power on or off to entire sections of the facility as needed, to save enrgy costs. One example of this is to use large commercial containers, such as those used on cargo ships, full of computers that can be quickly plugged in at a datacenter but only turned on and activated when demand or capacity is needed. All network switches, routers, server farms, and applications in the container or “datacenter in a pod” are configured via software for a particular customer or to increase capacity on demand only when needed. This isn’t science fiction; it is already happening today in some of the latest, most modern datacenters.

Cloud Service Brokering

As I detailed in Chapter 8, cloud service brokering refers to the role that an organization or cloud provider can take in delivering to customers a variety of cloud services from multiple providers. I expect a number of commercial community or public providers to launch cloud brokering services in the coming years in addition to large enterprise customers that are already planning this form of internal IT service aggregation.

Public sector government customers have shown a significant interest in cloud brokering. Many state and federal government customers already have the responsibility for consolidating IT purchasing — providing IT services or access to third-party vendors on behalf of entire government entity. Cloud brokering is simply a new style of delivering IT with responsibilities for evaluating which cloud providers are contracted, aggregating all service catalogs, consolidating billing across all providers, and governing the procurement and chargeback process to the consuming organizations.

A theme throughout this book is the future adoption of the hybrid cloud. Cloud brokering will be a method by which large organizations and public cloud providers deliver hybrid multiprovider functionality. Expect to see a significant rise in the popularity of and new services focused on hybrid clouds and cloud brokering very soon. Of course, as I stated earlier, many customers have a tendency to confuse the two delivery methods because each involves similar automated provisioning processes and technology.

Transforming Applications

As detailed in Chapter 4, every modernized datacenter or cloud provides, at a minimum, the basic infrastructure VM, storage, and network services. When you transform enterprise applications to the cloud, you can take advantage of the unique benefits of the cloud such as performance, elasticity, stability, and improved return on investment. Purchasing or deploying a cloud service is just the first step: application porting or redevelopment will become the long-term path to complete the transition to cloud.

There are not many applications that truly utilize the full potential of cloud computing at this time. Most of todays’ applications were created long before the cloud became popular and accepted, and it will take more time to reprogram or transform complex applications into the cloud. This transformation also means changing the applications so that they can split all processing and calculations into multiple threads or work streams to provide the dynamic scalability, performance, and reliability. Using multithreading, the application can spread its processing workload across numerous cloud compute devices, thus speeding up the application’s response to the customer. As the amount of data grows and application complexity grows, applications will need this capability to scale-out with cloud computers, forming clusters of thousands of computers. Many traditional enterprise IT functions and applications are expected to become cloud-based systems in the future. Inevitably, the cloud will become the first choice for application hosting and cloud native characteristics will be commonplace.

Continuous Application Development (Continuous Delivery)

The integration of continuous application development (also known as continuous delivery) automation, which is described in Chapter 4, will be a key differentiator for the leading public cloud providers and private cloud platforms. Most current cloud providers and management platforms support Development/Testing as a Service (Dev/Test) activities but do not truly implement the full continuous delivery model at this time. When continuous delivery is incorporated into the cloud portals and self-service control panels, Dev/Test will truly take advantage of cloud automation, software-defined networking, and drastically improve the application development lifecycle.

Application Publishing

Application publishing (detailed in Chapter 1) is expected to increase in popularity, particularly for mobile and laptop devices and users. Although virtual desktops have their place and purpose in the industry, it is application publishing of individual enterprise applications to a large number of users that will soon become a standard method for application management and distribution. Application publishing provides remote execution of applications that are hosted in the cloud, but the crucial capability is the ability to download the latest application version (in the background without the user’s knowledge) to each desktop or mobile device for use when offline from the network.

Open Source Application Development

The development of cloud-native applications is complicated because of the numerous programming languages and application platforms common in the industry. The latest trend, which is expected to greatly expand, is to use open source application development platforms to create multilanguage, multiplatform, portable cloud-native applications. Rather than supporting the traditional method of staging separate development environment for Java and another for .NET (to name just two platforms), new open source cloud development platforms such as Cloud Foundry support dozens of simultaneous languages and platforms. Add in cloud-based database/middleware platform automation and continuous application delivery automation tools, and the entire application development lifecycle will be truly revolutionized. Although these tools are individually available now, the combination of open source, cloud-native, multiplatform, continuous delivery, and automation is still a couple of years away from reality.

Further benefits of open source application development platforms are when you combine it with open source cloud platforms such as OpenStack. Using the same Cloud Foundry example, deployed on top of an OpenStack cloud platform, you now have an environment in which you can readily port any applications created within it to many other industry-standard cloud platform, application environment, or cloud provider that supports these open systems. The industry is a few years from this reality, but organizations should seriously consider using open source cloud platforms — certainly application development platforms — over proprietary systems and cloud providers to position themselves for future interoperability, cloud-native hybrid applications, and portability between cloud providers.

Application APIs

The industry is well on its way to implementing APIs into every aspect of cloud services, cloud management tools, and application development. The future of cloud computing, integration of applications, connectivity of applications to external services and other providers, and portability of cloud services between providers is all hinged on APIs. The industry is already seeing some of the world’s largest systems integrators, software developers, and cloud providers shift to industry or open source standards for APIs rather than proprietary APIs created by every provider (which is how the cloud industry first started). The quantity and level of sophistication of these APIs is where cloud computing will extend its reach into everyday business, consumer, and personal devices and applications. For more on this, see “The Internet of Things” later in the chapter.

Application Containers

One of the latest trends in cloud computing is application containers. This technology has actually been around for many years and is not specific to the cloud, but it has taken on new forms and popularity through open source products such as Docker. With the application container concept, you can “package” applications into self-contained executable programs that you can deploy onto multiple servers and even onto different OSs. These applications actually run in their own memory space and execute separately from one another because they share a common underlying application platform service, such as the Docker Engine in this example. This “Dockerized” application can run on any server or OS that has the Docker Engine, without needing unique versions or recompiling the app for every unique server or host, or cloud.

This concept of application containers might not be an entirely new technology; however, given the current and future demand for open source and hybrid clouds, expect Docker and similar application containerization tools to become even more popular.

Self-Service Administration and Control Panels

Future clouds will have a significantly enhanced focus on customer-facing self-service control panels — well beyond the basic service administration tools available in current clouds. Cloud management systems not only handle customer orders, orchestration, and provisioning functions of the cloud, but also provide a web-based portal from which customers can manage their cloud applications. Given the basic premise that cloud applications are hosted by a centralized provider (often a third-party), the administration of the applications also needs to be cloud enabled. Whether internal support personnel or a third-party contractor provides the application and day-to-day user administration functions, there are significant benefits to managing all of the cloud applications through a consolidated, consistent user interface or self-service control panel.

As mentioned in Chapter 7, most cloud management portals enable only limited IaaS-centric self-service capabilities, such as the ability to restart or reboot VMs. As the next generation of clouds focus on migrating more applications (not just IaaS), the cloud management portals will need to improve significantly in the area of self-service control panels because applications have significantly more settings and administrative tasks than fairly simple IaaS VMs.

With self-service control panels, you can define multiple roles for customer support staff and optionally end users, with each role assigned unique permissions and abilities to customize or manage applications. Customer support staff might perform tasks such as user account management, configuring applications, resetting passwords, changing mailbox features, adding new websites, enabling smartphones, starting and stopping VMs, or any other administrative functions the cloud provider allows. Or, the end user might be restricted to only changing her name, address, phone number, and password for an application.

The provider and/or customer administrators control what permissions or self-service capabilities are available for each role and users assigned to those roles. Many cloud platforms have a unified multitenant portal to which all customers can log on. Depending on the logon ID, the customer will only be able to see and manage their organization, their user accounts, and their applications; no customer can see or manipulate any other customer’s account. The cloud service provider support staff themselves also use the same management portal, but their logon ID permits more capabilities in the system. This does not mean that the cloud support personnel can see any customer private data; they just have the ability to make configuration changes to applications (not data) on behalf of the customer when requested. All changes made to services or configurations are fully logged for auditing purposes.

These are just a few examples of how customers can administer their applications and cloud service subscriptions. If the cloud provider does not have this capability, customers have no choice but to submit a traditional helpdesk support ticket and wait for the work to be done. This forces customers to place a call for every minor administrative task, which is neither cost effective nor a good customer experience.

Key Take-Away

An application self-service control panel is very often overlooked or not available from traditional IaaS-focused cloud providers. Just as you need to manage and configure VMs in an IaaS application (for which most cloud providers do offer a web portal), cloud consumers need SaaS management capabilities to centrally administer cloud-based applications. Future cloud deployments will have significantly more self-service administration capabilities to match the growing number of enterprise applications running to the cloud.

Big Data and Analytics

Big Data refers to the gathering of large amounts of data, often from various sources in the cloud, and storing this data into large databases. These databases and methods of collecting data are a perfect fit for a scalable, elastic cloud-hosted service — it provides customers with access to data that would have been impossible or impractical to gather or host within a legacy IT environment. It is when this data is processed through analytic tools that statistics, trends, and intuitive searching is accomplished — providing business intelligence that organizations have never had in such as useful form until now.

There are two primary categories of data: structured and unstructured. A database with defined fields (e.g., name, address, and phone number) is an example of structured data, whereas unstructured data might be emails or social media for which the actual data is rather random in form and content. Tools to analyze structured data have been available and used across multiple industries for many years. Unstructured data is the most common and fastest growing form of data, so analysis has been difficult. Previously, it could be performed only at a basic level, manually, by humans. When you combine a huge amount of data from multiple sources, both structured and unstructured, this is Big Data. Analytics is the process and tools that you use to work the data into usable views and reports.

Some real-world use cases for big data analytics include the following:

Retail products

You can perform real-time data gathering and create dashboard views of customer sentiment for a new product launch by gathering purchasing, social media, email, user forums, and customer service ticketing systems. Analytics tools can produce intelligent reports to manufacturers on customer acceptance or complaints almost immediately after a new product launch. You can use data trending to improve future products and customer service.

Security and intelligence

Audio, video, social media, email, or logged sensor data gathered from multiple sources can be analyzed to correlate (sometimes seemingly unrelated) events to proactively identify potential threats or criminal activity. The most advanced video processing, combined with analytics, can even detect potential terrorist threats, such as if someone enters a subway station with a briefcase and then later leaves another station without the briefcase. Government intelligence agencies worldwide are also known to use collected audio, email, and website data to identify potential threats or trends. Although this level of surveillance and analytics is usually reserved to intelligence agencies and the military today; it is only a matter of time before this technology is widely used by commercial organizations, or someday even end consumers.

Heathcare

Researchers can use long-term gathering of healthcare data from thousands or millions of patients to determine trends and patterns that no individual hospital, doctor, or research firm could gather and analyze manually or in isolation. Using real-time data gathering from personal wristwatches or medical devices embedded in patients, virus outbreaks, epidemics, and other events could be detected almost immediately.

It goes without saying that all of this structured and unstructured data gathering, along with the analysis output, can represent privacy issues. Ideally, the output data analysis would only represent an aggregated report and statistics without any personally identifiable information (PII). While it is relatively easy to implement PII data isolation and only show the consolidated statistics, the fact that the raw data is collected and stored online also means there is always a potential for data leakage due to security flaws or hackers.

Big Data and analytics are already here to stay in the industry. In the very near future, we will see huge advances in the sophistication of analytics tools and an explosion in the number and diversity of worldwide business consuming or subscribing to this business intelligence data. Everything from customer sentiment, products and prices at retail stores, social media reaction to events and television shows, to commuter traffic will generate data with ever-improving analytical tools producing intelligent reports to business (and consumers) via the cloud — even more exciting is that this will be done in near real time.

The Internet of Things

The Internet of Things (IoT) — or, as its sometimes called, the Internet of Everything — has recently been discussed in the industry. This essentially means that there will be a drastic increase in the number and type of devices that are interconnected to one another and to the cloud. The key to this IoT is that it will go well beyond business uses and eventually affect just about every aspect of human life, from healthcare to home automation to automobiles. Using APIs and improved standardization, many different devices from all manufacturers will be able to communicate. While I could write an entire book on this topic alone, I will focus instead on how cloud computing will soon be part of most daily tasks and devices you use.

The IoT, the Cloud, and your Home

We are already seeing home appliances such as televisions, digital video recorders (DVRs), and even refrigerators that communicate with one another via a wireless home network or the Internet. They also communicate via cable, satellite, and cloud-based content providers such as Netflix. Now take this even further into the home automation trend where a centralized computer (within the home or in the cloud) controls the lights, entertainment devices, appliances, heating and cooling, security sensors, and even the lock on the front door. All of these devices, and the entire home, can be controlled via your smartphone or by any computing device connected to the cloud. This gives it the ability to sense when you or a family member arrives home, automatically adjust the air conditioner, disarm the security system, turn on lights, and notify other family members of movement or activities sensed within the home.

One excellent example is the standard alarm clock that is a thing of the past. The future alarm clock (actually already available today) will be fully integrated into your home, detecting sunrise, adjusting room temperatures just before you wake up, slowly fading up the lights, opening window shades, and starting your preferred music playlist. Taking this alarm clock example even further, imagine just reaching over and pressing a button that automatically emails your office that you will be late or are sick and not coming to work today — shutting the lights and window shades so that you can go back to sleep.

Combine this automated home with the cloud and you get even more capabilities such as automatic updating of firmware in your appliances, unlimited music and video selections to download, visibility or awareness of home activities with an elderly family member across the country, and even synchronization of the home with your automobile. Using the alarm clock example again, imagine the home automation system waking you earlier than normal when abnormal traffic conditions exist on your normal commuter route. The cloud makes this and countless other scenarios possible when integrated into your home, automobile, and personal life.

The IoT and Cloud in Your Automobile

Most modern automobiles have internal software (called firmware) that runs everything from the engine management system to the radio or entertainment devices. Extending the reach of your automobile to the cloud and to other devices such as your home, smartphone, and even other vehicles will bring a new driving experience that you can customize and integrate with the rest of your daily routine.

Let’s look more closely at your cars firmware. Presently you must visit your local dealer or manufacturer to troubleshoot a problem or update the internal computer software. The newest trend — some could say started by Tesla Motors — is to design the cars with self-service or automatically upgradeable software as a core feature. With this functionality, the dealer can update the car with new features or improvements via the cloud even while you’re driving down the road. So the days of just having satellite or streaming radio to the car from the cloud are evolving to cars that are permanently connected to the Internet and cloud.

When your car is connected to the cloud, it will take on significant new features, customization, entertainment, navigation, and even engine monitoring and troubleshooting capabilities. Imagine your car dealer or manufacturer automatically monitoring your car’s performance, efficiency, or event logs and calling you when a problem is detected or even fixing the problem without you even noticing something was wrong in the first place. Now that is true customer service. Realize that this benefits the auto dealer and manufacturer, as well, with reduced visits to the service center and continuous improvements to their products by collecting and analyzing all the data from other customers of the same vehicle type. Yes, there are security and privacy concerns to be worked out, but we’ll leave that topic for another book.

Now, let’s take your cloud-connected automobile and integrate it with your home. Imagine that your home will sense that your vehicle has pulled into the driveway or garage and then automatically turns on lights or performs other customized tasks within your home. Take this a step further and let’s have your home computer’s music library automatically update wirelessly to your car so that you never have to put your music on a USB stick (or stream from your smartphone which drains the battery). Performance and efficiency data could also be synchronized automatically to your home computer for long-term trending and reports. You could also send navigation instructions or your favorite restaurants to your car, ready for your use via the onboard vehicle navigation system later. And, of course, real-time monitoring and tracking of your teenager as he drives: how fast is he going and where he is going is always popular (with parents at least...).

Although not yet a reality, one of the most exciting areas in the confluence of automobile and cloud is the interaction with other vehicles, other people, stoplights, even the road itself; that will be the next evolution for the auto industry. There are already applications for your smartphone with which you can see and interact with other drivers to determine how fast traffic is moving and redirect your navigation system to find the fastest route. Currently, these peer-to-peer applications are based on smartphones communicating with one another, but this will soon be a part of the automobile’s internal capabilities. Also consider the progress being made in autonomous vehicles that drive themselves by using continuous data from roads, sensors, and other nearby vehicles to help navigate and avoid one another.

The point to all of this is that the cloud is what enables all of these possibilities. Whether the cloud is used just as a communication channel (e.g., streaming music from the cloud), or as method of data transfer and interaction with other devices, it is the software APIs that makes possible the interaction with other cars and devices in the world, and that makes this a very exciting future.

Personal

Let’s not forget personal devices and how they will interact with the cloud in the near future. We already have smartphones connecting to the cloud, but the number and diversity of personal devices that also integrate with the cloud is about to explode. Watches are now being developed that monitor your pulse, blood pressure, oxygen saturation, and blood-sugar levels to name just a few. There are glasses that are now connected to the cloud, wristbands for fitness tracking, and even GPS tracking chips that you can embed in your pet. Key areas in which the cloud will change daily life include (but certainly are not limited to) the following:

Healthcare

Cloud-enabled monitoring devices will not only track the health of you or your family, but automatically notify your doctor or even emergency services during certain health-related events. Fitness tracking, today often using wristbands, is already commonplace. Given the vast amount of continuous data, your doctor will have an exponentially increased capability to analyze trends, see how medication is working, and even call you to come into the office before you know there is even a potential problem. These capabilities are even more effective and important for elderly family members — sharing information with doctors but also key family members if desired. Using Big Data and analytics, as described earlier, will revolutionize the healthcare industry when combined with the cloud and personal health monitoring.

Presence awareness

The ability to track family member locations anywhere in the world or just within your home or neighborhood will have a significant impact in your life. Just as you could embed a location-tracking device in your pet, you could track a person’s location by their wristwatch, smartphone, or potentially a human-ready form of embedded chip. Take this even further and use these personally worn devices to automatically open or unlock doors, start your car, turn on lights at home, or change music and digital pictures on the wall to your personal liking by sensing your presence — even as you move from room to room within your own home. Tracking this data and integrating with the cloud had almost infinite possibilities for convenience, lifestyle, and security.

Perhaps the scariest part of personal devices and cloud-based integration is the privacy. There will be privacy issues that we will all need to deal with and solve, but most of these technological capabilities will be optional for you and configurable in what data is held private versus what is shared to others (for instance, your doctor). One area you may not be aware of is the value of this personal health, activity, and presence awareness data to businesses. Businesses gathering and analyzing personal data at this level of detail can gain vast knowledge of habits and trends — personally or as a demographic. This data would be invaluable for companies to create and adjust their service or product offerings, safety, quality, and so on. This analytical data is not new; it is already common practice for most retail stores, healthcare providers, and even your Internet provider. When the data gets as detailed and personally identifiable as the wristwatch, embedded chips, and healthcare is where most people draw the line today on privacy. Ensuring privacy while still benefiting from these new technologies will be (already is) a significant challenge for personal interactions and the cloud.

Closing Comments

Throughout this book, I have analyzed lessons learned and best practices for cloud computing based on experience from early-adopter customers and industry cloud providers from 2010 until now. The cloud computing industry has matured at a pace even faster than anyone predicted. As I look back on the first generation of clouds, it is clear that some concepts now so critical to success were not well understood or even thought of back then.

As we study the lessons learned from the industry and first generation of cloud customers, I can clearly see new trends, technologies, and business processes that will be a focus for the next generation of the cloud. The future of the cloud is not just about technology or transforming legacy enterprise IT; it’s also about your business — prioritizing what business needs should be addressed first and potentially transitioned to the cloud. Using guidance provided in this book, organizations and business executives can take advantage of this knowledge to better plan, deploy, and complete their transition to cloud. It is no longer a matter of convincing you of the benefits of cloud computing; it is now time to consider when and how you will make the transition from enterprise IT to cloud-based IT service delivery.