Calculating the True Costs and Benefits of Media Analytics in the Cloud

Despite its reluctance to entrust the cloud with sensitive or mission critical tasks, the media industry is well on its way toward deploying the majority of its enterprise workloads in the cloud. The insight, based on data from 451 Research, points to three key benefits as drivers of the media industry’s move to the cloud.

As many media organizations know, a primary benefit associated with moving workflows and processing to the cloud is that this shift reduces or eliminates the need to maintain the on-premise infrastructure supporting that work. Freed from those responsibilities, IT staff can focus more time and energy on other activities that bring value to the business. Another primary benefit of the cloud is scalability. In delivering needed resources on a near-instant basis, the cloud allows the media organization to adapt with agility in the face of new challenges and opportunities.

The third benefit and driver noted by 451 Research is that of improving infrastructure total cost of ownership (TCO), which we have found to be a top concern of the executives determining business value of technologies and solutions, calculating their ROI, and establishing IT budgets for them.

It might seem that ROI calculations for the cloud would, by nature, be nebulous and difficult to pin down, but we at Decentrix have seen firsthand how the cloud both facilitates the reduction of capital investment and supports the alignment of business requirements with IT resource capacity. Our core technologies enable AI-driven media analytics on the ground, in the cloud, and in hybrid configurations. Using accurate, objective measurements to evaluate our customers’ cost savings in employing cloud-based media analytics, we have been able to assess the significant that migrating such analytics from the ground to cloud has on TCO.

Assessing the True Economic Impact of the Cloud Transition

As a media enterprise grows, it moves through various iterations and configurations of storage and processing systems. The cloud dramatically simplifies the tasks of managing and orchestrating a path forward through this evolutionary process. Standing up and then continually reconfiguring the local datacenter hardware required for media analytics requires much more time and a much greater investment than does pursuing similar technical and functional progress in the cloud.

In understanding the actual costs of relying on a cloud platform rather than an existing on-premise business configuration, it is critical to understand two primary classes of cost: solution processing costs (SPC) and variable environmental costs (VEC).

SPC have to do with equipment and licensing costs required to achieve the required business outcome within service-level agreements (SLA). For the purposes of evaluation, they include not only the selection of machines supporting business operations, but also the processing times for those machines. For most organizations, analytics requires about 10 TB of storage capacity, and an on-premise hardware solution will provide a useful life of three years before an upgrade is necessary.

VEC concern the costs relating specifically to the datacenter and compute environment. These costs reflect the priority — and associated level of investment — that the media enterprise places on factors such as high data availability (HA) and disaster recovery (DR), and they therefore are widely variable across different businesses. Given the variability of VEC and the difficulty of parsing and pro-rating them across the overall IT infrastructure, these costs are hard to calculate specifically. They can, however, be estimated based on current investment in HA and DR.

Assuming that the analytics platform itself costs the same whether on premise or in the cloud, it is possible to use SPC and VEC to determine what efficiencies and savings can be achieved in the cloud.

Analyzing the Costs of Cloud-Based Media Analytics

Table 1. An analysis of basic SPC related to the media analytics configuration, see above, shows that an on-premise data center solution is about three times more expensive than a cloud-based model. Click to enlarge.

Table 1. An analysis of basic SPC related to the media analytics configuration, see above, shows that an on-premise data center solution is about three times more expensive than a cloud-based model. Click to enlarge.

SPC for on-premise media analytics can be normalized to a 24-hour processing day, as shown in Table 1, and broken out in terms of server type and licenses required, along with total investment period (three years), cost per day, and percentage of use applied to analytics. Similar numbers can be established for the functional equivalent in cloud-based storage and compute resources. Cloud-based models typically lack direct costs for licensing and instead build those costs into hourly usage rates.

Analysis of basic SPC related to the media analytics configuration clearly shows that an on-premise approach is 2.9 times more expensive than a cloud-based model. If an HA/DR environment is a must, then the media enterprise must establish a redundant facility in a separate geographic location. As a result, the costs of an on-premise deployment double, rising to 5.8 times a cloud-based deployment. The media enterprise also must consider the fact that, unlike a cloud-based solution, an on-premise deployment requires a large upfront capital investment in equipment — servers, storage, and networking equipment — and, in a handful of years, the cost of upgrading that equipment.

Due to the fixed and inflexible nature of local infrastructure, the media organization’s efficiency with respect to resource use is lower in an on-premise solution than in the cloud. (If the organization has opted for an HA/DR model, then this efficiency is halved.) The overall costs of deployment can be mitigated if locally installed equipment can be used for other purposes when not engaged in provided computing resources for the media analytics solution. That said, it can be difficult for an organization using on-premises hardware to align its capacity with its actual requirements. More often, the organization will provision and pay for the infrastructure essential to address peak demand, regardless of how frequently this capacity is needed. Insufficient investment in resources can lead to system downtime, and unintentional overprovisioning of infrastructure can result in excessive spending on unnecessary capacity.

Because the cloud model depends on economies of scale and mixed workloads, it yields efficiencies and cost saving only when the cloud implementation is managed to the economics of the media organizations. If basic SPC are extrapolated across 24 hours and compared with an equivalent on-premise solution, then the on-premise solution will yield a cost advantage of 1.9 times that of the cloud solution.

Special Considerations for Cloud Configurations

The flexibility offered by the cloud is among its key benefits. With respect to implementation of a cloud-based media analytics solution, this flexibility yields several specific results. For one, it’s more cost-effective to add future data sets. Rather than expand on-premise infrastructure with large monolithic servers, the media enterprise can incrementally add smaller (and more economical) servers and services.

In acting as a scalable platform for a growing population of analytics users, the cloud makes it easy to increase reporting capabilities on demand. Both processing power and storage resources can be scaled up and down in alignment with demand. The cloud also simplifies geographic distribution of data for DR; backups of data and configurations can automatically be stored across different cloud datacenters to enable data protection and quick restoration in the event of disaster. Likewise, cloud vendors’ use of server clusters helps to maintain the high availability of data despite any drive failures.

The Cloud is a Win-Win

This analysis shows that when implemented correctly — in alignment with actual technical and business requirements — a cloud-based media analytics system offers the greatest benefits in terms of TCO, and also with respect to maintaining agile operations. Using SPC and VEC to assess its requirements and calculate actual costs, the media enterprise can make an informed move to the cloud and significantly reduce the overall cost of using an advanced media analytics solution.

Travis Baker serves as the director of support engineering for Decentrix Cloud Services.

Travis Baker serves as the director of support engineering for Decentrix Cloud Services.

Let us know what you think…

Log-in or Register for free to post comments…

You might also like...

IBC 2019 Cloudy, Streaming and 8K

In this thought-provoking missive, Gary Olson delivers his predictions and insights for IBC 2019.

In Live Channel Creation Video Quality Monitoring Is Critical!

There is always a cacophony of noise around the big tradeshows; at NAB this year Telestream made its voice clearly heard with the first showcase of OptiQ Channel, a new and radically different one-click channel creation live service for cloud…

Taming The Virtualized Beast

Without doubt, virtualization is a key technological evolution focus and it will empower many broadcast and media organizations to work differently, more efficiently and more profitably.

AI – ML – DL – The Differences

Artificial Intelligence is more than just one element. In this article, we look at and describe the many parts AI encompasses.

Cyber Security - A DAM Issue

After visiting the recent Henry Stewart DAM (Digital Asset Management) conference in New York, Gary Olson asked some very difficult questions of Cloud vendors regarding security. Their responses may surprise you.