Cloud computing has enabled a radical shift in the way modern enterprises manage their businesses and data. Offering unmatched computing speed, reliable performance, and enhanced productivity, cloud technology also promises organizations the benefits of reduced CapEx. Firms no longer need to invest in bulky and expensive hardware and software that were required to manage onsite data centers in the past.
However, there are hidden cost implications of cloud computing in the form of constantly surging OpEx, and this often results in large, unexpected invoices that leave enterprises grappling to optimize their IT budgets. In our earlier paper on Managing Cloud Costs, we discussed the 3 main problem areas including cloud cost visibility and transparency, cloud cost optimization, and cloud cost management. This paper, in continuance, discusses the economics of traditional IT models vs. cloud computing, public cloud in particular. It explores how organizations can optimize their CapEx and OpEx with public cloud to better allocate budgets and stay on course on their IT spends.
The prelude - Economics of traditional IT models
Traditionally, enterprises have looked at IT as a necessary investment to ensure smooth functioning of organizational systems. Over the last three decades, IT has evolved to play a pivotal role in prioritizing business decisions and investments with the computerization of various business systems including the systems of record, engagement, and finally, systems of insight. These systems have provided a solid foundation for organizations by ensuring maximization of performance and throughput designed for peak loads. However, with constantly evolving technology, CapEx has continued to rise steadily, accumulating a lot of inefficiencies in IT systems hosted in data centers.
All this while, OpEx across organizations was driven down drastically with increased automation of IT systems aided by centralized co-ordination and management of virtual machines. However, this also resulted in increased complexity of systems, further fueling automation and more CapEx investment. While financial engineering did play a major role in converting some of the CapEx into OpEx through various financial models, customers were still locked into multi-year contracts that required significant initial investments and were expensive.
About 2010 – 2015, the advent of cloud technology promised enterprises the advantage of on-demand, large scale data storage infrastructure and computing power that was available with no financial lock-in. The biggest advantage of the cloud is that it offers enterprises economies of scale, as they only need to pay for the services they use. This is perceived to lower both CapEx and OpEx and enable better infrastructure management. More organizations than ever before are migrating to the cloud. Consider this: According to IDC, worldwide public cloud services spending will more than double by 2023[1].
Inherent challenges of the cloud model
Despite the promise of financial freedom and ability to select IaaS (micro, nano, large, extra-large instances), PaaS, containers or cloud native services (for eg., serverless) based on the enterprise’s needs, cloud technology has some inherent limitations in the areas of security, compliance adherence and the organization’s ability to control their IT environment. Ironically, all of these issues are silently contributing to yet another challenge –cost management. Let us explore these limitations in more detail:
Growth of hyperscale computing
Innovation driven by hyperscalers provided cost benefits, but also increased complexity of operations. The benefits of cloud outweighed concerns around outages and increased surface for attacks. Organizations have and continue to invest heavily in strategic and consulting engagements that provide better maturity models, encouraging them to adopt cloud. Increasingly, more and more enterprises are technologically locked into the engagement, with the slow realization that there are unexpected and hidden costs riding on these cloud investments.
Lack of control
Enterprises end up storing their data with another party, outside of their physical and organizational control, trusting that their data will be secure and maintained well. Many organizations find this lack of in-house control of the data servers unsettling. This cautious approach to the cloud has resulted in shadow IT steadily gaining prominence, silently fueling unforeseen and unplanned expenditure.
The problem of plenty
Another important challenge that cloud poses is that of choice. Customers continue to be flummoxed by the different cloud options available and the services each entails, including the different sizes, types and reusable components from the providers. The cloud marketplace has evolved exponentially over the last few years - IaaS, PaaS, SaaS, serverless compute functions, cloud native services, readily consumable business services, customizable services/functions. All of these have opened up a plethora of adoption models making the right choice of cloud services a very difficult one for the customer.
Containerization
Servers, virtual machines and cloud instances have now matured into containers that best utilize underlying resources to run microservices. Running containers on cloud leveraging multi-cloud orchestration delivers cloud agnostic workloads that could scale in an agile manner. Monolithic workloads are being rewritten for cloud – to run on containers and to be able to scale across cloud or move across clouds.
However, the complexity of estimating the cost of running these microservices is probably the toughest challenge for technologists. Containers can scale infinitely depending on the load – running only parts of the code that is relevant as a microservice and scaling back when done. However, what organizations do not realize, and service provides often fail to stress upon is this - this architecture has multiple moving parts that doesn’t seem to be contained in a predictable cost model.
Spiraling OpEx
One of the biggest advantages of the cloud is the shift from CapEx to a pay-as-you-go model. Much like utilities billing, customers pay only for what they use, when they use it. However, while this proposition is an easy call to make for start-ups, it merits considerable thought and careful planning for large enterprises in the long run.
Often, the overall price tag of cloud services ends up being much higher than anticipated for large scale enterprises. 50% of enterprises spend more than $1.2 million on cloud services annually [2]. Most large organizations, because of the amount of data and processes they run, end up overprovisioning for their services, and maintain a hybrid cloud model, incurring large OpEx costs that were unplanned for in their budgets. If they are not careful, these costs can spin out of control, doing away with the biggest advantage of the cloud.
Balancing costs with technology innovation
Despite the challenges, cloud is driving innovation heavily across industries. The public cloud service market is expected to reach $623.3 billion by 2023 worldwide.[3] Boundaries are blurring and technology is enabling new avenues of growth for traditional businesses. Data is now seamlessly leveraged across domains, and services are hyper-customized to users resulting in empowered end users. The growing importance of digitization is now reflected in new organizational roles such as ‘Chief Digital Officer’. The ability to deliver custom content, lean and agile processes enabled by DevOps, microservices on containers, PaaS, and other cloud native services will determine next gen digital business. Efficient cloud systems of the future will be driven by the ability to predict the need, availability of efficient levers to scale quickly, and agility to deliver appropriately.
However, as IT management models evolve, cost imperatives are being significantly impacted. Organizations, particularly the larger ones, must ensure that they continuously monitor and improve their OpEx even as upfront CapEx seems to be under control. As mentioned in How to Manage Cloud Costs, organizations must have a thorough understanding of their CSP’s billing and invoice structure. This is particularly crucial for organizations operating on hybrid platforms, as hidden and unforeseen costs can significantly impact overall budgeting. Efficiency is guaranteed when performance and throughput are maximized while costs are kept under control. Enterprises must make a conscious effort to correlate business impact with ongoing spends to measure efficiency and take corrective action where required. Best practices to continuously monitor OpEx costs on cloud include:
Businesses must make use of available information that pinpoints the inefficiencies in the systems and continuously work to improve their OpEx margins. IT must be chartered with the task of maximizing output and reducing inefficiencies in the system even as innovation continues to multiply the speed at which business happens.
Seamless migration to cloud-native state
Cloud service providers and hyperscalers must enable organizations to better estimate their consumption with a proactive quota and budget management system.
Wipro’s holistic extensible platform ‘BoundaryLess Enterprise (BLE)’ enables enterprises to rapidly adapt to the cloud-native digital world by seamlessly integrating and operating multi-cloud services. BLE provides organizations with a ready-to-use, plug and play platform that delivers cloud services for multiple stakeholders customized to their needs. The solution manages hybrid cloud-native applications lifecycle on CI-CD through DevOps extending to edge services.
Most interestingly, BLE offers automation of workflows to rapidly provision, control, secure and terminate cloud services. It ties in the subscription to the user and the service – so that enterprises can meter their cloud consumption needs and stay on top of their costs. The optimization module in BLE helps organizations estimate their bill of IT, monitor real time consumption and correlate billing data. It further provides organizations with recommendations on optimization, enabling them to operate on the cloud efficiently by improving performance and reducing costs.
Click here to know more about Wipro’s BoundaryLess Enterprise offering. For any query reach out to CIS marketing at marketing.cis@wipro.com
References
[1] https://www.idc.com/getdoc.jsp?containerId=prUS45340719
[2] https://hostingtribunal.com/blog/cloud-adoption-statistics/#gref
[3] https://hostingtribunal.com/blog/cloud-adoption-statistics/#gref
Girish Jagajampi
Cloud Practice Lead, Cloud & Infrastructure Services, Wipro Ltd.
Girish has more than 20 years of experience in the IT industry across software development, datacenter services, DC consolidation & transformations, virtualization and cloud services. He leads the engineering efforts to build and enhance cloud offerings and is focused on delivering solutions that define next-gen services.
Connect with him at girish.jagajampi@wipro.com
Click here to download your copy of State of IT infrastructure report 2020