The history and evolution of software supply chains

Today, software is ubiquitous and has been for more than half a century. From the early EDVAC system to the current Apple Watch, every computing device has needed software. However, software development as a practice was not as popular in the 1950s as it is today. Not only has the speed at which technology is developed and shipped increased manifold, development itself has been optimized and made simpler with the use of tools. The seminal shift caused to this increased popularity, speed, and efficiency can be attributed to the advent of software supply chains, often known as DevOps.

The supply chain management term was coined in 1983, but long before that, supply chains have been revolutionizing industries including manufacturing, consumer packaged goods (CPG), healthcare, and now e-commerce, among others. Less knowingly so, it has also revolutionized the technology sector. It is not often that we associate supply chain processes to the development of technology, at least not as much as we correlate the impact of technology and improvement of supply chain processes. Understandably so, over the last three decades it has often been a business prerogative to digitize supply chains and use technology to optimize processes and reduce cost. While the previous generation of business leaders had prioritized technology for supply chain, in the last decade we have seen business leaders in large technology companies prioritize supply chain processes to improve the productization of technology. This prioritization has been so implicit operationally that several millennials and Gen Z that represent the majority of the booming technology sector today do not even realize that they are often part of an efficient supply chain.

The software supply chain is a process of developing, distributing, and maintaining software products. It involves various stages, such as designing, coding, testing, and releasing software, as well as managing updates, and security patches. The supply chain often includes multiple parties, such as software developers, third-party vendors, and end-users. As the name suggests, often the well-known supply chain processes are at play in making the production line more efficient, higher quality, and low cost for software product development.

Although at the surface individual supply chains look different through the vast supplier structures and distribution, there are often core similarities. For example, specifically the software supply chain is similar to the auto supply chain in the following ways:

• Tiered supply chain structure. Both the auto supply chain and software supply chain have a tiered structure, where different suppliers provide various components that are then integrated into the final product. In the auto supply chain, for example, suppliers provide parts such as engines, transmissions, and interiors, which are then assembled by the automaker. Similarly, in the software supply chain, different vendors may provide software components such as libraries, frameworks, and APIs, which are then integrated by software developers to create the final product.

• Quality control. Both supply chains require rigorous quality control processes to ensure that the final product meets the necessary standards. In the auto supply chain, for example, each supplier must meet certain specifications and undergo testing to ensure that their components are reliable and meet safety standards. In the software supply chain, vendors must similarly provide documentation and undergo testing to ensure that their software is secure, reliable, and compatible with other components.

• Just-in-time delivery. Both supply chains rely on just-in-time delivery of components to minimize waste and reduce inventory costs. In the auto supply chain, for example, suppliers may use just-in-time delivery to provide components to the automaker precisely when they are needed for assembly, reducing the need for warehousing and storage. Similarly, in the software supply chain, developers may use just-in-time delivery of software components to avoid building up a large backlog of untested or unnecessary code (often known as technical debt).

• Complex dependencies. Both supply chains have complex dependencies between different components, which can create risks if a supplier fails to deliver on time or produces faulty components. In the auto supply chain, for example, a delay or defect in a single component can disrupt the entire production process and delay the delivery of finished cars. Similarly, in the software supply chain, a vulnerability or bug in a single component can create security risks or cause the entire software system to fail.

Interestingly, a lot of terminologies used in software supply chain processes are adopted from the traditional supply chain ecosystem. Methodologies such as waterfall, Agile (Scrum, Kanban, and others), iterative-waterfall, and spiral models are all derived from the traditional supply chains. Similarly, DevOps is made up of workflows and concepts such as continuous integration (CI), continuous delivery (CD), defect density, among others; all of which have been assimilated into the technology product creation process from the traditional supply chain fields. Finally, Toyota’s lean supply chain models and just-in-time manufacturing have also been heavily adopted by the technology startups of the present day.

While it is true that the advent of technology innovations and the resulting software products have revolutionized the traditional supply chains in various industries, it is undeniable that the technological innovations would not be in their current form without the supply chain processes fueling it. Although it may be appropriate to use the chicken-and-egg parabolic metaphor to explain this situation, it is often hard to establish which is of greater significance: technology or supply chain. Whatever be the right answer to this, it is clear, that we see a truly symbiotic relationship that is activating a rising spiral of economic development.

About the author:

Saikat Banerjee, is an engineering leader at Amazon Web Services (AWS), Inc., the cloud computing arm of Amazon. He specializes in the development of large systems and running teams delivering engineering innovations. He is a technologist with over a decade of experience and has worked on large scale systems digitizing supply chains across various industries. Saikat holds a Bachelor’s degree in Computer Science and Engineering and a Master’s degree in Engineering from Massachusetts Institute of Technology (MIT) with a focus on Supply Chain Management, Sustainability, and AI.

SC
MR