What Is Cloud Cost Optimization? Strategy & Best Practices

Kevin Bogusch | Oracle Senior Competitive Analyst | January 22, 2024

Anyone who’s owned a home is familiar with the challenge of maintaining a clean garage. You start with a neat environment and what feels like limitless space for storage, a workbench, and maybe even some exercise equipment. But soon a few boxes go here, a few more go there, and the area becomes a chaotic maze with little room to walk, much less park a car.

Cloud computing can present the same challenge. As businesses take advantage of the cloud’s scalability and flexibility to provision servers and computing instances in minutes, cloud environments can quickly become cluttered with unused or underutilized resources, adding unnecessary complexity and chewing up IT budgets. After all, though cloud computing offers customers the benefit of paying for only the cloud resources they choose, cloud service providers (CSPs) may still charge customers whether they use those resources or not.

This is where cloud cost optimization comes in. Its goal: Make cloud environments more efficient and less complex, and ultimately less costly. This article examines how this process works, why it can be difficult to implement, and which best practices to follow to optimize cloud costs successfully.

What Is Cloud Cost Optimization?

Cloud cost optimization is the process of reducing the overall costs of cloud computing services while maintaining or enhancing performance. The goal of cloud cost optimization is to align costs with actual needs without compromising on service quality or performance, typically by limiting expenses such as overprovisioned resources, unused instances, or inefficient architecture. It’s a balancing act between keeping costs down and providing the appropriate cloud resources to maintain peak performance, fuel growth, and ensure compliance and data security.

Cloud cost optimization is also a dynamic process because cloud workload requirements constantly evolve, as do cloud pricing and service options. As a result, cloud cost optimization requires detailed metrics, analytics, and automated tools.

In general, cloud cost optimization involves two core initiatives.

  • Intelligent procurement of cloud services: One benefit of cloud computing is that it allows authorized employees to purchase cloud services quickly to meet demand. But, despite their best intentions, IT teams can inadvertently sign up for more resources than they need. That’s why strong governance policies for cloud purchasing are necessary to help ensure that companies get the most value from their cloud investments. Governance involves setting clear budgets and reviewing billing information; leveraging cost-saving opportunities, such as discounts for volume purchasing and paying in advance; and monitoring cost anomalies to identify and address unexpected spikes or inefficiencies.
  • Optimization of cloud capacity: The ease and speed with which IT teams can purchase and provision cloud resources can sometimes have costly, unexpected consequences. If companies don’t have visibility into resource usage, for example, they may unintentionally overprovision cloud resources by purchasing more server or storage capacity than they need, leading to idle or unused capacity. Teams can also forget to shut down resources they no longer require. Companies can mitigate this by continuously monitoring usage with cloud monitoring and automation tools and conducting regular optimization reviews. This process, known as “rightsizing,” helps ensure that the most cost-efficient cloud resources are allocated to each workload or application. In addition, companies should monitor for unused cloud software subscriptions, which can also add unnecessary costs.

Key Takeaways

  • A crucial benefit of cloud computing is the ability to add servers, storage, and networking capacity quickly and easily to respond to usage demands.
  • Cloud pricing has become increasingly complicated, which can cause companies to inadvertently overspend on unnecessary resources.
  • Cloud cost optimization helps companies control cloud costs and improve budgeting, forecasting, and IT performance.
  • Best practices for cloud cost optimization include setting strict budgets and using automated tools to identify and adjust cloud resources in the moment.

Cloud Cost Optimization Explained

In the early days of cloud computing, companies eagerly took advantage of the cloud’s scalability, flexibility, and easy provisioning, often without fully understanding costs. But as cloud adoption soared, pricing and service models became more complex, resulting in underutilized cloud resources and unexpected cost overruns for many businesses.

And so cloud cost optimization was born. Early attempts at cloud cost optimization involved manually monitoring usage and adjusting resource allocation, but continued cloud growth made this process a challenge. For example, cloud providers began to offer almost unlimited options for instance sizes for workloads. In addition to server size, IT teams had to select options for memory, databases, computing power, graphics, storage capacity, and data transfer speed, among other variables. With so many factors to consider, choosing the correct size instance for workloads became difficult, with many companies unwittingly purchasing more capacity than they needed.

To help customers avoid unnecessary expenditures, cloud providers have started to offer comprehensive cost management tools that provide insights into resource utilization, cost breakdowns, and recommendations for optimization. In addition to these cost management tools, many companies have developed strategies and best practices to maximize their cloud investments. These include using automation to scale resources up and down as needed, identifying discount opportunities with cloud providers, and continuously monitoring and adjusting their active cloud services.

Many companies also take a structured approach to cloud cost optimization by assigning dedicated team members to oversee the process. That responsibility can span roles, including cloud architects, cloud operations managers, cloud financial analysts, and cloud cost engineers. More commonly, companies create a cloud governance board with multiple stakeholders. Also referred to as financial operations or FinOps, this team is charged with developing and implementing best practices for cost management, maintenance, bulk shutdowns of unused resources, and other cost-cutting procedures.

Why Is Cloud Cost Optimization So Important?

Cost control is obviously a primary goal of cloud cost optimization, but it’s not the only reason this process is important. Cloud cost optimization also addresses challenges with cloud performance optimization and security, while providing other benefits.

Here are the most important reasons to implement a cloud cost optimization strategy.

  • Higher cost savings: Estimates of wasted cloud spending range from worrisome to downright alarming. In a 2023 Flexera survey of global cloud decision makers, respondents reported wasting an estimated 28% of their public cloud spend. A focus on cloud cost optimization best practices and policies helps create a culture of cost awareness, teaching teams to consider the return on any cloud spending. This enables companies to base future cloud purchasing decisions on solid data rather than hunches. According to McKinsey Digital, the results of cloud cost optimization can be dramatic, allowing technology leaders to “quickly cut as much as 15 to 25% of the costs of their cloud programs while preserving their value-generating capabilities.”
  • Improved efficiency: Underused or idle resources, poor application optimization, and mismanaged cloud resources can add significant costs to cloud operations. For example, an ecommerce company might pay a flat monthly fee to run several maximum server loads 24/7 during peak buying seasons. During nonpeak hours, however, those servers might run at 10% capacity, meaning 90% of the company’s spend goes to waste. Rightsizing and autoscaling tools can help companies identify and merge underutilized or overprovisioned resources to reduce costs while increasing application performance. Those cost savings can then be reallocated to other parts of the business that deliver greater ROI.
  • Smarter budgeting: Cloud cost optimization strategies and budgeting have a mutually beneficial relationship. Developing and communicating detailed cloud budgets helps maintain cost controls. Likewise, cloud cost optimization strategies help companies forecast future cloud spending and manage budgets more effectively via historical usage analysis and improved cost visibility. Cost monitoring and reporting tools can provide insights into resource usage, cost outliers, and how different services or workloads contribute to overall cloud costs. This creates a more accurate baseline for budgeting and forecasting.
  • Enhanced performance: Cloud cost optimization also has important implications for application performance. As noted earlier, CSPs offer seemingly limitless options for customers to set up cloud instances, which often leads companies to purchase more services than they need. A by-product of cloud cost optimization is a better understanding of the distinct requirements of different workloads. The process provides companies with operational metrics that help them select performance thresholds for each workload more accurately, which can enhance processing times and user experiences.
  • Reduced security risks: Cloud cost optimization isn’t the most important component of a comprehensive cloud security program, but it can certainly help the cause. For example, by merging or eliminating overprovisioned and underutilized resources, companies can reduce their attack surfaces. In addition, the increased visibility that comes from monitoring cloud usage can help companies identify potential security threats. Cloud cost optimization tools—such as autoscaling or infrastructure as code (IaC), which automates cloud provisioning—can also help enforce security controls consistently and reduce the risk of misconfigurations.
  • Improved business continuity: Cloud cost optimization encourages companies to distribute resources across multiple availability regions to enhance resiliency. This process can help improve business continuity by reducing the risk of downtime and minimizing the impact of disruptions. In the event of a disruption, cloud cost optimization tools for IaC, for example, can speed up the deployment and testing of infrastructure and applications to get businesses back online quickly.
  • Greater sustainability: In general, most things that reduce waste promote sustainability. By rightsizing their cloud services and merging underutilized resources, companies can allocate cloud resources more efficiently, which can help minimize their energy consumption and reduce their carbon footprints.

Why Is Controlling Cloud Costs So Difficult?

Keeping cloud costs under control can feel like an exercise in futility. That’s because many of the benefits of the cloud, such as self-service and limitlessly scalable resources, can be a blessing and a curse if not properly managed. Complex cloud pricing models are often the root of the problem.

For example, software-as-a-service (SaaS) pricing is typically based on the number of cloud subscriptions a company purchases, requiring companies to carefully monitor their subscriptions to ensure they don’t go unused. Meanwhile, infrastructure-as-a-service (IaaS) models are often based on the amount of computing, networking, and storage capacity a company reserves on a CSP’s platform each month, creating even more complexity.

In a decentralized cloud environment, IT teams may find themselves with a newfound ability to make immediate decisions regarding new cloud resources. As more teams across an organization take advantage of this ability, the costs can quickly add up, particularly if no one monitors whether new resources are necessary or how long they’re required. Autoscaling features offered by some cloud providers can help control costs, but they’re not a panacea. Companies still need to establish clear autoscaling policies that specify scaling triggers based on performance metrics and establish minimum and maximum scaling limits.

The complexity of cloud billing can also compound the challenges of cloud cost optimization. The problem: All those countless cloud configuration options can have their own respective pricing model. As a result, the average cloud bill contains hundreds if not thousands of lines of data. As CSPs add new features and pricing structures, the complexity increases further. This is especially true for companies that use multiple CSPs, each with its own billing terminology. In most cases, the task of understanding and allocating each line from a cloud bill falls to a finance professional who likely doesn’t have the training or experience to interpret the charges. As a result, they will likely be unable to advise IT teams on how to optimize spending.

Get Started with Cloud Cost Optimization

Cloud cost optimization is a daily practice. Unlike accounting, for example, where monthly or quarterly reporting requirements dictate when companies must ramp up their activities to meet established deadlines, cloud cost optimization is proactive and constant. Ongoing cloud innovation and shifting organizational priorities make careful attention to detail vital for cloud cost optimization. With this in mind, the sooner companies can build a standing group of diverse cloud stakeholders to oversee cloud costs and policies, the easier ongoing cost control will be.

8 Cloud Cost Optimization Strategies and Best Practices

Clear policies for purchasing and implementing cloud resources are foundational to best practices for cloud cost optimization. Once these policies are in place, companies can embed them into cloud workflows to automate the process of real-time discovery and timely response.

The following eight best practices can help companies establish cost discipline for cloud spending:

Eight best practices can help companies establish cost discipline for cloud spending
Applying best practices such as reviewing billing, setting budgets, identifying unused resources, or taking advantage of discounts can help businesses establish cost discipline in their cloud spending.
  1. Review pricing and billing information for anomalies: Interpreting a CSP’s often lengthy and granular cloud bill is easier when companies take the time to zero in on specific areas of high expenditure. For example, it’s important to fully understand charges for compute, storage, and value-adding managed services, such as database services, because these tend to make up the bulk of cloud costs.

    Most CSPs offer cost management tools that can identify trends, anomalies, and cost drivers before customer spending gets out of control. In addition, recent cloud cost management innovations have incorporated machine learning to detect unusual patterns in cloud usage and costs. Heat-mapping tools visualize peaks and valleys in demand, which can help companies determine when to shut down cloud services to save money. Most cost management tools also allow customers to set up alerts to notify them when costs or usage exceed predefined limits.

    In addition, companies can take advantage of tagging strategies to manage expenditures. CSPs allow customers to tag items on cloud bills—by department or project, for example—to help segment costs into customized categories to assess the ROI of specific cloud investments.

  2. Set budgets: Once companies understand cloud billing and usage patterns, they can more easily budget for future spending and avoid unexpected costs. Unlike on-premises IT expenditures, which require high and often unknown up-front investments, cloud spending involves monthly subscriptions with fees based on usage that can vary from month to month. Because of this, companies must implement governance policies that align IT costs and performance without throttling the agility that cloud computing offers. Doing so requires in-depth collaboration between finance and IT. Companies, therefore, should strongly consider including IT team members, such as developers, systems operators, and security professionals in their cloud budgeting discussions.

  3. Take advantage of cloud native design: Companies have several options to choose from when considering a move to the cloud. Lifting and shifting on-premises environments to the cloud allows them to quickly migrate their on-premises applications without modification—but it comes with trade-offs. Since most legacy applications weren’t designed for the cloud, companies can’t make use of key benefits, such as managed services for maintenance, patching, and updates. Legacy apps also tend to use cloud resources inefficiently, adding to cloud costs.

    By contrast, cloud native applications are designed with efficiency in mind. By designing or using cloud native apps, companies can take advantage of managed services as well as critical cost optimization tools such as autoscaling. Following a cloud native development strategy, however, might incur additional up-front costs as companies may need to train staff on an entirely new development methodology.

  4. Identify unused or idle resources: With so many instance options available, cloud administrators can inadvertently opt for too much computing power. In addition, developers can easily create compute instances, load balancers, storage volumes, and other cloud resources as needed—but they may forget to deprovision these resources when a project ends. Depending on their payment plans, companies may be charged for unused or idle resources, ultimately paying for more than they need.

    As a best practice, use cloud management tools from CSPs or third-party providers to track use, identify idle or underused resources, and find potential savings. For example, Brazilian furniture retailer Tok&Stok uses a monitoring tool from Oracle to achieve greater autonomy over its cloud resources, reducing costs incurred by idle servers. The tool automatically adjusts compute capacity to scale in line with the demands of daytime operations and peaks versus nighttime or weekend operations. Cloud platforms can also trigger alerts when usage falls below predetermined levels, and idle and underutilized resources can be merged into fewer instances to reduce costs and increase application performance.

  5. Rightsize cloud services: Once they’ve identified underutilized cloud resources, companies can modify their workloads for usage, size, and cost efficiency. The process of rightsizing involves analyzing usage patterns and performance metrics for applications and workloads, typically using cloud cost management tools. Through regular monitoring and analysis, companies can identify mismanaged cloud resources and realign them with the needs of each workload. For example, an instance optimized for memory might ultimately end up running more compute-bound tasks, either because it was overprovisioned or because the purpose of the application changed over time. In this case, a business could save a lot of money by rightsizing to a compute-optimized instance.

    Companies can accelerate their rightsizing efforts with automation tools, such as autoscaling and IaC technologies, that trigger immediate action based on continuous analysis. Most CSPs offer autoscaling tools that automatically add to or reduce a customer’s server instances and storage according to that customer’s predefined parameters. A retailer can automatically add cloud servers to handle peak holiday shopping workloads, for example, then automatically scale back as demand declines. This autoscaling helps ensure that customers don’t pay for unused server instances. As another example, cloud CRM vendor Star CRM uses autoscaling within its Oracle Cloud Infrastructure (OCI) environment to adjust dynamically for peak periods. The company pays for additional compute capacity only when necessary.

    In addition, IaC automates the provisioning of servers based on customer input, so new server instances are automatically optimized with the right amount of compute and storage capacity for their specific workloads. This helps companies avoid wasting money by unintentionally misconfiguring their cloud resources.

  6. Look into discounts and savings plans: Most cloud providers offer discounts to help businesses reduce their cloud spending, though not all workloads will be eligible. The following three discount programs may result in significant savings:
  • Reserved instances: Reserved instances (RIs) can deliver substantial pricing discounts if companies commit to using specific instance types over a defined period, typically one to three years. Although RIs are unlikely to be effective for unpredictable workloads, companies can save up to an advertised 75% compared with on-demand pricing for stable and predictable workloads, such as mission-critical apps that run continuously. Because RIs are generally billed up front, companies should have a clear understanding of their long-term usage patterns to determine accurate commitments.

    Cloud providers also offer savings plans that, like RIs, are based on prepaid commitments to use cloud resources over time. Savings plans are commitments to spending, generally measured per hour, regardless of instance type or region. Savings plans can offer more flexibility than RIs, which are essentially commitments to capacity levels and specific instance types. As a result, savings plans make more sense for companies that expect a certain amount of cloud spending but whose needs are likely to change.

  • Spot instances: Think of spot instances as cloud auctions conducted by CSPs to sell unused inventory. While these last-minute resources can come at tremendous savings—up to an advertised 90% off on-demand pricing—they also come with a few caveats. Spot instance discounts fluctuate depending on availability and demand. Thus, there’s no way to predict if or when spot instances will become available or whether a bid will be accepted. Even if a company is successful in purchasing a spot instance, the instance often comes with limitations. For example, CSPs can interrupt spot instances with little notice if the costs exceed the company’s maximum bid price or availability becomes limited. This makes spot instances unsuited to mission-critical or time-sensitive workloads. Ideally, they should be used for noncritical workloads that can withstand interruptions or are designed to automatically shift to alternate resources in such an event, known as a “graceful shutdown.”

  • Volume discounts: Many CSPs offer volume discounts for larger customers based on tiered pricing defined by the provider. It’s a simple concept: Per-unit pricing declines as customers use more of a service. Some companies prefer to use multiple cloud providers to avoid getting locked into one vendor, but volume discounts may make it worth consolidating specific workloads with a single provider. Take, for example, a company that uses three CSPs, spending $700,000 with one and $200,000 each with another two. If a provider offers a volume discount for spending more than $1 million, the company may save by consolidating.

  1. Limit data transfer fees: CSPs often charge customers to migrate data between different regions and availability zones or across different services within their cloud ecosystems. For companies that frequently move or replicate data across regions or services, data transfer fees can add up quickly. CSPs charge customers for data ingress (data coming into the cloud) and data egress (data moving out of the cloud), with data egress typically incurring higher costs. This can occur if a company has inefficient data retrieval processes, is overly reliant on transferring data for routine operations, or has insufficient data lifecycle management practices. Other factors that impact data transfer fees include redundant transfers—when independent teams migrate the same data—and a lack of data deduplication and compression.

  2. Manage, organize, communicate, and educate: Cost management tools and best practices are key to establishing a cloud cost optimization process that centers on communication, collaboration, and education. The task of building a culture of cost awareness around cloud spending should fall to a FinOps team comprised of IT, finance, and project staff. Together, these resources must establish processes, policies, and frameworks to review, monitor, and control spending across the organization. The FinOps team should further encourage communication and cross-functional alignment on cost optimization by bringing together different lines of business to promote companywide engagement, accountability, and knowledge.

    The FinOps team is responsible for establishing two-way communication channels to share best practices, strategies, and new cost management tools while allowing for companywide feedback. Companies may also consider holding seminars and training sessions to educate stakeholders on crucial topics, as well as incorporating cloud cost awareness training into their onboarding processes.

What to Know Before Moving to the Cloud

Successful cloud cost optimization doesn’t start after cloud implementation. It begins before migration. As companies build a case for moving to the cloud—and as they navigate the provider selection process—they have a unique opportunity to establish a culture of cost awareness.

Building a collaborative FinOps team is the first step. This group should oversee cost control and all policies related to how cloud technology gets purchased and implemented throughout the organization. Here’s how the FinOps team can get a head start on crafting a cost-effective cloud strategy.

  • Review service level agreements (SLAs): Provider SLAs reflect their commitment to several important performance variables for their products, including uptime, performance metrics, support response times, and data availability. A cloud provider’s SLA must align with customer requirements. Companies that require high availability for critical applications, for example, should confirm that their chosen provider offers guaranteed uptime to meet their needs.
  • Assess total cost of ownership (TCO): Cloud computing costs extend beyond tangible expenses, such as subscription fees. A wide range of intangibles must be accounted for as well, such as the impact of downtime or slow computing speeds on productivity or potential sales. TCO analysis tallies all the tangible and intangible costs of implementing, operating, and maintaining a cloud environment over a specific period to help companies compare vendors and calculate accurate budgets and ROI.
  • Evaluate provider scalability options: The scalability of the cloud is one of its biggest selling points. To get the most from their cloud investments, companies should carefully assess the options each CSP offers. For example, cloud providers that offer demand-based autoscaling and adjustable compute or storage capacities can help companies ensure their cloud resources aren’t wasted.
  • Prioritize seamless integrations: Any new cloud provider should fit cleanly into a company’s existing IT architecture. A cloud-based customer relationship management (CRM) system, for example, should integrate as seamlessly as possible with existing customer data or other apps that rely on CRM data. Compatibility among systems can reduce costs and minimize disruption.
  • Understand security policies: Cloud computing has proved to be capable of being just as secure as on-premises computing—if not more so. However, companies must ensure their cloud providers meet their specific security requirements. Different industries may have different regulatory and compliance guidelines, such as the Health Insurance Portability and Accountability Act (HIPAA) in the United States and Europe’s General Data Protection Regulation (GDPR). To avoid costly penalties, companies should thoroughly review a cloud provider’s compliance information, paying close attention to their encryption protocols, access controls, vulnerability management, and data privacy measures.
  • Assess backup and disaster recovery strategies: A CSP with strong backup and disaster recovery capabilities can save its clients significant time, money, and anguish in the event of an outage. When assessing providers, consider their backup frequency and data retention policies. Cloud providers with IaC and autoscaling capabilities—automated to provision new servers and scale resources as needed—can also help customers get back up and running quickly after an outage.
  • Train and develop employees: Investments in cloud skills development provide IT staff and employees with the knowledge and ability to operate cost-effectively in a cloud environment. Making certification and continuous learning programs available can also help teams adapt to new cloud innovations and emerging best practices.

Get Optimized. Cut Cloud Costs with Oracle.

As well as offering flexible pay-as-you-go and subscription-based pricing models, Oracle Cloud Infrastructure (OCI) features a suite of cost management tools to help proactively manage and modify cloud usage and spending, making it a cost-efficient cloud solution. Oracle cost management tools focus on the following five critical areas of cloud cost optimization:

  • Organization: Organization tools help companies link and manage multiple tenancies in OCI to limit usage, for example, in the event of overconsumption. They also support critical tagging functions that allow companies to segment and allocate costs by project, region, or any other category.
  • Reporting and analysis: Companies can visualize and track spending based on customized parameters and build usage reports for audits or invoice reconciliation. Oracle’s cost reporting tool also analyzes invoice line items at resource-level granularity.
  • Governance and control: OCI’s governance model provides a structure to design and improve cloud security, address compliance, and reduce costs. Once parameters are set, companies can more easily enforce policies with tools that, for example, put quotas on consumption.
  • Budgeting and forecasting: Oracle’s tools emphasize cost visibility by allowing customers to set companywide thresholds for spending, with automatic alerts when users approach their limits. Oracle also offers a cost estimator tool so prospective customers can gauge monthly usage and costs before committing.
  • Optimization and cost reduction: In addition to tracking usage, Oracle’s optimization tools proactively identify underutilized cloud resources, offering recommendations and quickly rightsizing based on company guidance. Oracle also offers reward tools for users who successfully manage cloud spending.

Cloud cost optimization is as much a cultural initiative as it is a technical one, though cost management tools will always play a key role. By combining those tools with clear policies and effective communication strategies, companies can maintain control over their cloud spending and reduce the likelihood of exceeding their budgets.

5 Actions CFOs Can Take Now to Fuel Profitability and Growth

Discover game-changing tactics, from driving transformation with AI to fine-tuning your M&A strategy and embracing collaborative leadership, to help you navigate today’s challenges—and tomorrow’s—and thrive.

Cloud Cost Optimization FAQs

How much does a cloud setup cost?

Cloud implementation costs can vary greatly based on several factors, including the size of the company’s data stores, its computing requirements, the number of applications migrated to the cloud environment, and the amount of data transfer required. Typically, cloud implementation costs include setup fees from a cloud service provider as well as per-user subscription fees for cloud applications, the cost of internal and external resources to plan and manage the implementation, data migration costs, and employee training costs. Depending on the scope of the project, additional full-time IT resources may be required as well.

What are the different types of cloud costs?

In addition to the initial implementation costs described above, cloud costs include monthly fees for subscriptions and the consumption of cloud resources such as server, storage, and networking capacity. Cloud service providers may have different pricing structures, but businesses typically are charged monthly for the number of servers used, the amount of data stored in the cloud, data transfers between the cloud and external sources, database services, technical support, and additional services such as analytics or security.

Why is the cloud so expensive?

While some companies may consider cloud computing expensive, many others view the cloud as the most cost-efficient computing model. Compared with on-premises computing, for example, the cloud eliminates the up-front expenses of hardware and software, as these are handled by cloud service and application providers. While cloud computing includes monthly subscription and consumption costs that don’t apply in an on-premises computing model, many companies find the flexibility of the cloud and the ability to scale resources based on demand make it easier to keep costs under control.

Is the cloud really worth it?

Determining whether investments in cloud computing are worth it comes down to the individual needs of each company. Many organizations have found that cloud computing is well worth the investment, and spending trends on the public cloud bear that out. In general, cloud computing customers have cited cost-effectiveness, scalability, and on-demand services as top benefits of the cloud.