DevOps Practice

Challenges of Designing API-Driven Experiences

As SaaS and mobile application development–and cross-software integration and communication–continue to become increasingly important for organizations across all industries to remain competitive today, APIs play a critical role. In fact, APIs dominate digital experiences today, with an average of 220 new APIs published every month, representing a 30% increase over the previous four years.

As more organizations shift to an API-first development strategy to further drive innovation, partnerships and rich end user experiences, the proliferation of demand and consumption of APIs can bring its own challenges, if not properly executed from the get-go. Amidst the ever-increasing rush to develop and publish an API to remain competitive, we see several common issues arise–and often only after the API is rolled out–that negatively impact integrations, future partnership opportunities and end user experiences. Read on for a look at these common challenges and how to address them before you hit publish on your next API.

Infrastructure Costs

One of the biggest challenges when it comes to API design is the ability to manage costs, in particular infrastructure costs. From managing multiple gateway servers and instances, to building an entire API management program from the ground up, we see many common infrastructure approaches that organizations deploy today when designing APIs that lead to unnecessary and often unwieldy expenses.

Critical to avoiding spiraling infrastructure costs is leveraging a single gateway to deploy, govern, secure and deliver global API traffic across various data centers. By using this approach, organizations can reduce man-hours spent managing and synchronizing servers; increase reliability by eliminating the need to manage multiple gateway instances; govern APIs without needing to build and maintain a separate API management platform; and eliminate regional gateway replication.

By considering an API gateway at the onset of a design project to ensure API centric operations (such as authentication, authorization and throttling) are happening at the edge, organizations can proactively control costs and avoid financial surprises.

Scale

To a developer, there’s nothing worse than having a wildly successful API that can’t scale to meet the demand.

Addressing scalability early on in the process can help define both early adoption, future success and the lifespan of the API. However, organizations commonly overlook the need to manage scalability in several ways–including planning for surges in API traffic and managing the volume of API requests consumers can make–which can lead to damaged user experience.

Developers should look to a combination of load testing, authentication, throttling, quota management and API caching at the edge to improve and predict traffic, while preventing infrastructure from being overwhelmed by requests to ensure availability and reliability for consumers. Building in load testing early on can help a developer determine the traffic volume the application can withstand when surges in requests occur.

Quota management can help enforce business service level agreements and limit the number of API requests that a partner is allowed to make. When it comes to API caching, some good places to start include: any resource accessible via HTTP GET, static data, immutable responses, infrequently altered or predictable responses and frequently requested data.

Security

While the commercial value of API development is clear, the reality is APIs can come with great cybersecurity risk. For one, APIs can provide a glimpse into the back-end of an application implementation and even the database it is connected to, providing hackers with new avenues for attacks. As the communication bridge between multiple applications, APIs left unprotected can also increase an organization’s attack surface and expose it to downtime and malicious attacks, including unintended misuse by legitimate users.

Critical to protecting APIs is incorporating security at the infrastructure level with a multi-pronged approach that includes leveraging an API gateway to easily validate, authorize and control the access of legitimate API consumers (and block legitimate users), and endpoint protection against malicious traffic. Additionally, rate limiting–which puts caps on the number of requests per minute or second that API consumers can make–can prevent adversaries from overloading origin in an attempt to bring down API infrastructure in the form of a DDoS attack.

Organizational Decentralization

As speed and agility increase exponentially and API development evolves from microservices to nanoservices with individual team members owning what feels like ever-shrinking components, it’s easy to see how teams can organizationally splinter. While autonomy is important to making the development process flow quickly, decentralization is creating major governance and communication issues within many organizations today.

This lack of governance creates its own challenges related to configuration visibility, changes and duplications across implementations which could lead to wasted resources, data leakage and more. Organizations should consider a single gateway that can expose APIs across multiple, distinct implementations but also provide suitable isolation so as to not affect the performance of other APIs when requests occur, while also providing separate control pages and the ability to set up distinct authorization and access rules for each API. The idea is the APIs are managed in the same decentralized way that is required to meet speed and agility requirements under a centralized infrastructure and governance policies.

Strategic API design is not a simple task, but it is absolutely critical to organizations that desire long-term API success. By seeking cost-effective infrastructure, tapping tools to ensure scalability, building security protocols, practices and policies, and taking steps to more centrally manage the API design process early on will ensure longevity for the API, driving strong partnerships and, ultimately, providing better user experiences.

 

This article was co-authored by Anthony Larkin, director of product marketing at Akamai. Anthony leads the go-to-market strategy for Akamai’s performance product portfolio. For more than 11 years at Akamai he has been dedicated to helping businesses remove barriers and unlock their potential to better engage users through web and mobile applications.

Sid Phadkar

Sid Phadkar

Sid is a product manager at Akamai, focused on enabling delightful experiences for their developer base. He is currently focused on making Akamai an organic component of their user’s continuous integration workflows as well as making Akamai the go-to platform for any API traffic needs for customers. As a PM, Sid loves to understand the why behind things. He is passionate about making decisions informed by customer stories and data. Prior to joining Akamai, Sid spent a few years consulting tech companies in optimizing development lifecycles and a few PM years at Dell EMC launching the company’s first-ever subscription-based product offering aimed at hybrid cloud data centers. Sid holds a computer science degree from UT Dallas and an MBA from Duke University (he has particularly strong opinions on Duke basketball). In his spare time, Sid can be found trying to make an impression in local pickup soccer leagues around Boston.

Recent Posts

The Role of AI in Securing Software and Data Supply Chains

Expect attacks on the open source software supply chain to accelerate, with attackers automating attacks in common open source software…

1 hour ago

Exploring Low/No-Code Platforms, GenAI, Copilots and Code Generators

The emergence of low/no-code platforms is challenging traditional notions of coding expertise. Gone are the days when coding was an…

20 hours ago

Datadog DevSecOps Report Shines Spotlight on Java Security Issues

Datadog today published a State of DevSecOps report that finds 90% of Java services running in a production environment are…

2 days ago

OpenSSF warns of Open Source Social Engineering Threats

Linux dodged a bullet. If the XZ exploit had gone undiscovered for only a few more weeks, millions of Linux…

2 days ago

Auto Reply

We're going to send email messages that say, "Hope this finds you in a well" and see if anybody notices.

2 days ago

From CEO Alan Shimel: Futurum Group Acquires Techstrong Group

I am happy and proud to announce with Daniel Newman, CEO of Futurum Group, an agreement under which Futurum has…

2 days ago