Could Serverless Architecture Be the Key to Cleaning Up the Enterprise Document Mess?

BrandPost By Soumya Chatterjee
Jan 30, 2020
Cloud ComputingIT Leadership

istock 542718630
Credit: miko315/istock

Document storage may be among the first reasons a company might consider the benefits of cloud computing in its enterprise transformation. But a comprehensive system to manage those documents is among the last – or even wholly overlooked – parts of many companies’ migration to the cloud. Now, however, thanks to the cloud computing model known as Serverless, it may be the right time to address the sprawl of documents across an enterprise, inside of which may be locked the insights that are truly differentiating for an organization.

How we got here, part 1: The Documents

Ever since the concepts of value were abstracted as “money” and agreements as “contracts,” business has run on documents. Desks of clerks gave way to typing pools and graphics departments, which were themselves replaced by the means for every employee to create, collaborate, and distribute documents. Analysis by McKinsey of IDC data revealed that in employees now spend 28% of their time reading and answering email, a fifth of their time searching and gathering information, and only 14% of their days communicating and collaborating internally.

The rapid rise in the amount of content a business produces has aligned naturally alongside the rapid digitization of business operations and functions. The TCS 2020 CIO Study of more than 1,000 corporate technology leaders revealed that, on average across 11 industries, 48% of business activities – from the products and services they sold to their legal and human resources departments – were digitized as of today.

The answers varied by industry and revenue, of course. For example, while banking and financial services industry CIOs said that nearly two-thirds (65%) of their finance and accounting operations were digitally enabled, only 41% of retail CIOs said so. The legal departments at companies with $50 billion or more in annual revenues are approximately 57% digital today, whereas at companies with between $500 million and $1 billion in revenue, only about 45% of the lawyers’ work is digitally enabled.

In every area of business operations, however, digitization is expected to only increase between now and 2030, with the expectation that, on average, nearly three-quarters (73%) of business activities would be digitized by 2030. And a large part of that work will be in how all those documents all those parts of the company produce, use, maintain, update, distribute and discard during the tenure of an employee, the terms of an agreement, and the lifecycle of a product or a project.

How we got here, part 2: The Servers

The abstraction of functions and processes in business, enabling new and more flexible revenue models, has followed the abstractions that make computing more powerful and more readily available to enterprises of every size. Today, a start-up can assemble a supply chain, its distribution channels, and even sales and support functions nearly as quickly as it can deploy a worldwide and infinitely scalable presence and operations, thanks to cloud computing.

Abstraction and virtualization in computing is what has enabled the cloud to replace the data center, which itself was an abstraction to replicate the purposes of a single server. Today, a company’s developers and operations teams might be working to deploy new capabilities on “the server” – but in fact, the server is more of a concept than an actual, individual assembly of hardware and operating system software they could point to.

Yet even in the cloud environment, application development has largely been a server-based exercise, with little to no distinction given between the “server” that stores the data and the “server” that runs the code – the server (abstracted into the cloud) is still the central computing resource.

What “Serverless” means for enterprises

“Serverless” is called such because it breaks the convention of servers, making the resource itself transparent to the development and operations teams while allowing optimal coupling of application function execution and compute resources. Serverless cloud computing, as offered by TCS partnering with AWS, treats the storage of an application’s data as a service and the application’s operations as a service. In fact, it is this latter emphasis – computing functions as a service, or FaaS – by which serverless computing is also known.

At first glance, the benefits of serverless computing appear to be in cost-savings, and for large enterprises with millions and millions more computer operations every minute, these alone could be considerable. Instead of maintaining cloud capacity to keep a “server” always running in order to operate an application’s commands as well as to store all of an application’s data, storage can be its own service – with pricing dimensions around availability, durability, quantity, performance, recovery, etc. — and the runtime of the application can be another service, with the enterprise paying only for the code that actually executes, and not for the unused potential to run all the code all the time that the traditional server model of computing requires.

But the true advantages that Serverless architectures offer an enterprise extend far beyond the cost savings. By removing the hardware considerations normally tethering engineers, application developers can use agile principles to build and rapidly deploy innovative new features and functionality without giving any thought to provisioning, capacity, OS patches, or maintenance. In other words – with a serverless architecture, innovations can automatically scale, remain reliable, and provide rapid global availability in the most efficient manner.

What serverless could mean for document management

As important as documents may be to its functioning and success, an enterprise’s “document management system” is typically a mix of ad hoc practices and imposed attempts to rein in an unwieldy, overgrown wilderness of data. Documents are most often scattered across digital and physical forms and each likely presents multiple vectors for security breaches. Moving this sprawling mess to the cloud may remove immediate constraints of on-premise servers, storage, and data centers, but an undifferentiated stack of documents is no more helpful to the enterprise than if they were locked away on the original separate hard drives inside respective applications.

Meanwhile, as the appetite for storage grows, modern storage “server” applications (in the cloud, but still always running in case it needs to process any application functions and retrieve and deliver its stored data) may merely extend the document sprawl, now only virtualized with unlimited storage resources. This may provide capacity and remove physical dependencies and constraints but does not address accessibility and organizational requirements to unlock the inherent value of the data in these documents.

A serverless document management system as designed by TCS using platform capabilities offered by AWS addresses this challenge from several angles to make further and faster enterprise transformation possible. By rethinking the functions needed to manage documents, a company can store and retrieve documents in the most effective and efficient way with a pay-as-you-use model for the application’s operation, separate from the storage costs.

An interactive dashboard allows users to manage documents with a few button clicks, monitoring document access and providing contextual choices for the next action in a workflow. Based on user needs and AI-driven rule exceptions, AWS Lambda runs the code to transition documents from Amazon S3 Standard to S3 Infrequent Access buckets or Amazon S3 Reduced Redundancy Storage – choices determined by business rules to facilitate access to the data while controlling storage costs. In order to fulfill statutory and regulatory requirements, the system will also automatically transition documents to S3 Glacier for the required 7- or 10-year retention period.

One major benefit of a document management system is a full indexing of repository and document contents, but this activity and access to the index data – along with the wide variety of access permissions across an enterprise’s documents – are among the vectors malicious actors might try to use to breach security. By contrast, in the serverless system designed by TCS, base security controls are enhanced at each step of indexing and access using distributed ledger technology (such as blockchain), strengthening the system against unauthorized data modification and malign access.

Such an approach – Serverless, leveraging the capabilities of AWS cloud offerings such as AWS Lambda, Amazon API Gateway, and Amazon S3 storage along with other services, may be deployed across an entire enterprise. As well, due to the inherently distributed nature of the application – as additional departments and needs become digitized and require a cloud-native solution to their particular opaque document sprawl these may be brought online easily, and further additional uses and enhancements are innovated and deployed simply by invoking new code, all without the need for system-wide downtimes affecting current users.

Advantages of using serverless to solve document glut

The benefits long promised by document management systems – findability, true “knowledge management,” and leveraging the aggregate of employee expertise to realize new opportunities – is now realistically available for an entire enterprise regardless of size, due to the scalability, reliability and lower cost of Serverless services. Some specific features a serverless document management system provides include:

  • Intelligent cost-savings on storage and processing, drastically reducing billing charges for cloud applications, and ability to demonstrate accurate usage and cost patterns for enterprise data
  • Action-oriented intelligence as a pluggable solution to optimize storage expenses
  • API/service-based integration with the wider digital ecosystem, which can use the features of document management that are purposely “exposed” – perhaps even leading to new sources of revenue for the enterprise
  • Easily integrated AI-driven solutions for document processing, which can guide users on actions they should take to optimize storage choices for varying frequencies of document access
  • A pay-per-use model for application functions, whereby a company pays only for its actual operations and not the capacity to run hypothetical workloads
  • Strengthened security and end-to-end visibility of data integrity is enabled by distributed ledger (e.g., blockchain) technology

Perhaps most importantly – with the advent of Serverless functions-as-a-service architectures, the enterprise can be positioned for further digitization and transformation while finally coming to terms with the volume of documents it will continue to produce and increasingly need to make use of.

Learn today how TCS can help you unlock exponential business value with this approach.