API Management scalability
Jackie Zhu 1100007DBS firstname.lastname@example.org | | Tags:  api scalability apim redbooks
0 Comments | 1,374 Visits
Andrew Das is a Technical Pre-sales Leader at IBM, responsible for selling WebSphere DataPower appliances, IBM MessageSight and IBM API Management. He believes these technologies play a vital role in forming the underpinnings for Internet of Things and has been very active in developing assets and though leadership in this space. He currently supports the Banking and Financial Services industry, focussing on some of the largest IBM accounts. Andrew holds a Bachelors degree in Computer Science and a Masters in Business Administration.
The popularity of APIs as the primary interface for Mobile, Web and Embedded device applications has been confirmed by wide-scale adoption across multiple industries . APIs have become the de-facto standard due to their simplicity and self-service capabilities for accessing or manipulating digital resources hosted in the public, or enterprise domain.
Companies are now increasingly looking to outsource innovation by utilizing 3rd party cloud services, or build their on-premise solutions. While there are pros and cons to either approach, companies who take the latter approach initially struggle with defining a good scalability strategy for hosting a resilient, efficient and low latency environment. In the past, companies had greater control over services they provided on the internet. These public services were predominantly web browser based traffic when it came to public consumption. After a decade of experience scaling web infrastructures, this became a well-understood practice. Web services also became dominant during the same period; although the ecosystem of users was typically limited to hundreds, or thousands of users within a more controlled environment.
With the advent of the API economy, many of the same practices from the Web era applied but with a significant twist. The users could be in the millions accessing services through a variety of different devices. For example, a web browser user had a lot more patience in viewing a website, where several seconds were tolerated between page loads. With the rise of the mobile user, this experience was not at all acceptable. Mobile apps are now expected to provide an instantaneous experience. Delays arising from accessing APIs, or worse not being accessible at all could impact the entire mobile user experience, prompting the user to immediately search for alternative apps from a competitor. This could have devastating consequences for the company offering the app through lost revenue and market share.
Scaling an API Management solution is a combination of understanding the target market and the underlying product capabilities. There are certainly other infrastructure components such as the network, hardware, application servers and so on that have a strong bearing on a scalable design. But my focus is primarily on the API Management solution.
In the upcoming IBM API Management Redbooks publication, I have shared some widely accepted best practices for sizing and capacity planning an API Management solution. The chapter on Scalability is written in two parts. The first part covers scalability best practices that can be applied to any API Management solution, while the second part marries the best practices with IBM product specific capabilities.
Once the book is published, I will notify the readers of its availability, in a separate post and Tweet. So keep an eye out as I believe this is a vital topic that many can benefit from.
For IBM API Management related blog posts, see:
For IBM API Management Redbooks publication, see: