|By Alistair Farquharson, CTO, SOA Software||
|March 31, 2014 08:45 AM EDT||
APIs are transforming businesses to extend the reach of their information systems and data. In technical terms, APIs are rather finite, allowing two software applications to exchange data and execute procedures. Yet their business impact is so great that API management requires a far from finite thought process. A "soup to nuts" approach is needed to help companies determine how to achieve the most business success with APIs.
An API enables applications to access back-end systems using lightweight, standards-based protocols such as REST and JSON. For example, if you were in food distribution, the actual business of soup and nuts, you could develop an API that lets mobile app developers tap directly into your ERP system. They could build consumer apps for recipes that showed ingredients that you had in stock. Grocery stores could build apps that gave their shoppers bulk discounts on large orders and so forth. The API extends your food distribution business into customer realms where you've never played before. This is fantastic. But it also presents a number of challenges to the management of the business.
An API Is a Product
Whether you like it or not, your API is basically a software product. It should be managed like one. As external parties connect with your API, they will reasonably expect the kind of professional communication and seamless updates that they would find with any web-based software business. If you are a mobile app developer who is accustomed to tapping into Amazon.com's API or any number of comparable interfaces, you will expect the same kind of experience when you connect with a corporate API.
Perhaps more significantly, the API represents an investment of resources. As such, it should serve a well-thought-out and thorough business strategy just like any software product. This means managing its lifecycle. The software lifecycle starts with planning, continues through development and operation, and ends with retirement and replacement. Ideally each phase of the lifecycle should be managed and monitored for optimal results. Certainly better API lifecycle management will foster more satisfied developers and partner communities.
Planning and Building Your APIs
Sometimes an API just is. Someone in the IT department created it for some reason that has since been forgotten, and there it is. It may or may not be great, but you have it. Should you use it "as is" or wait? This is more common than you might imagine. If you are approaching your APIs as if they were products, you would wait and work through a plan first. You have to get together with business and technology stakeholders and determine the business purpose of the APIs that you will be building. Now, this begs the question: Who exactly is this "you" we are referring to? That is one of your first planning challenges. Who is going to own the API business products that will extend the business into new spheres? So first you have to figure out who owns the API, and often it will be a collection of people that represent different parts of a business. Then the next planning task is to understand the cost/benefit outcomes for the business and intended users.
Chances are there will be more than one API developed and within each API there will be a host of different features to be developed. You will have to devise a priority for the rollout of new features. For instance, with the recipe app, the first release of the API might include simple searching and browsing. The second release might enable users to order products for delivery. The third release could let users pay for orders with credits cards and so forth. Not all of these features can be perfected at once.
Once the API is up, the organization has to be in place to support it. It is necessary to think through the empty seats that might need to be filled so that users of your API will feel as if they are connected to a responsive, living business. For example, if you are planning to invite developers from Europe to connect with your API, you need to have a technical point of contact for them as well as documentation that will be meaningful to them, perhaps translated into different languages. You will have to structure your business to support and manage the API.
Running the API: Protect, Secure, Manage
Your APIs are doorways into your business systems. That's great, but it is also a bit scary. APIs need strong, coherent security and management. If you have a handful of APIs, you will be able to stay on top of their availability, security and provisioning by hand. As your API program grows and it will if you are doing it right, you will likely find that using an API platform becomes a best practice - simply because it can take care of the serious work of API management. Effectively running an API through a platform involves the following:
- Support non-functional requirements - These include message protocol handling, security policy, authentication and authorization, etc.
- Manage provisioning and access control for apps - This means the selective provisioning of API access. Corporate APIs are not like their consumer counterparts where you want millions of users. With a corporate API, controlled access is where the users are the most valuable, not the most numerous. The API platform should enable selective provisioning that is still highly automated and light in terms of administrative load. Unfettered usage can bring a host of problems. Traffic and load management can get strained. Worse is the possible need to add extra server instances of costly business software just to satisfy API-driven demand. If the increased load is not generating revenue, satisfying it with more instances will be a waste of money.
- Monetization and control - An API might be a profit center itself. For example, a food distribution company API could be monetized from selling access to data about sales trends in the food industry. This scenario requires the ability to license access to the API.
- Provide API monitoring - APIs are just like any other piece of enterprise IT that is monitored for its system health, response times, and availability. In some cases, uptime may not be a big issue. However when your customers and partners are connecting to an API, you will want to know whether it is up, down, or running slowly. The API platform should provide monitoring functionality as well as failover for APIs that go down.
Sharing Your APIs: Publish, Support, Syndicate
APIs succeed when they are shared. The API platform can help create a marketplace where developers can discover your APIs and request permission to use them. The marketplace can be internal to your company or set up for external relationships. Whether it is a developer portal or something comparable, the platform should provide the below sharing functionality:
- Enable you to interact with and recognize your API developers. In some cases, this process can be set up on a self-service basis to allow it to scale without a major resource investment on your part.
- Facilitate the creation of great documentation about your API and how to use it.
- Make testing against your API as easy as possible.
- Monetize your API to assist in future cost benefit analysis.
Analyze Your API Program: Measure, Report, Iterate
Measurement of program results is in the DNA of most good IT managers. The API platform can help make this process as simple as possible. You will likely want to measure and report on the success of your program across the whole API lifecycle. Lifecycle and results are usually linked. For instance, if you see adoption of the recipe API rise as new features are added and new versions are introduced, that is an important finding to determine the payoff of earlier investments in the program and guide future direction. The right API measuring tools can help you drive improvement back into the planning stage. It is a never-ending cycle.
The Platform Approach to API Lifecycle Management
The execution of an API program that manages APIs like products across their full lifecycles requires a combination of organization and technology. The technology alone cannot make it happen. The organization cannot do it alone without the proper tools. In our experience working with many large enterprises, the best practice is to match the API owners with a platform approach to lifecycle management. The platform can be a complete offering, such as our own SOA Software API Management Solution, or it can be built in a variety of other ways. What's important is to recognize that the people, process and platform need to work together to effect comprehensive API lifecycle management. This will ensure the success of the API program.
Ideally lifecycle management will be baked into the structure of the platform tooling, allowing for automated approvals and workflow for each stage of the API lifecycle. This means having the ability to align work streams related to API lifecycle, such as costing, product management, documentation and legal. People involved in running the API program can thus work together efficiently. The API development capability itself should have API modeling, templates, versioning and change management, and impact analysis. These features serve the goal of managing the full API lifecycle without an untenable resource investment.
API management typically works using a proxy that can be managed and monitored more easily than the API itself. For multiple APIs, the best approach is a "gateway" or collective proxy with added integration and mediation capabilities including:
- Process Composition/Orchestration.
- Security, including AU/AZ, attack prevention, and protecting your systems from abuse.
- Caching and paging.
- Supporting multiple mobile app platforms.
- Managing quality of service (QoS).
Exposing APIs to your systems involves building a community. Developers whom you don't know will be writing code to access your data through APIs. For the process to work, they need to feel as if they are connecting with real people. This can be achieved in part with self-service and automation and actual human involvement. A developer portal is essential for the following functions:
- Self-service community to promote innovation and lower support costs.
- Interactive documentation to increase adoption and encourage experimentation with your APIs.
- App provisioning so you can gain visibility and control over the apps that are hitting your APIs.
- Integrated testing to speed up learning.
- Analytics to provide feedback and measure success.
You have your soup. You want the nuts. There is a path to dessert. However all this requires a mature and profitable API program. Our recommendation is to think about what you want to accomplish with APIs from the vantage points of productization and lifecycle. Building APIs puts you into the business of creating software products. While you may not see the API as a product, it is one. And, like any software product, it needs to be managed across its entire lifecycle. From planning through running and sharing, the API has to be monitored and secured. The work of managing it needs to be handled efficiently so that the program can be financially and strategically beneficial to the business. The best practice to ensure all of these outcomes is to use an API management platform to automate the handling of all of your APIs as they progress across their lifecycles. That will get you the nuts. The alternative is to go nuts, and that's not what anyone wants.
"We help to transform an organization and their operations and make them more efficient, more agile, and more nimble to move into the cloud or to move between cloud providers and create an agnostic tool set," noted Jeremy Steinert, DevOps Services Practice Lead at WSM International, in this SYS-CON.tv interview at @DevOpsSummit, held June 9-11, 2015, at the Javits Center in New York City.
Jul. 2, 2015 04:13 PM EDT Reads: 652
The basic integration architecture, as defined by ESBs, hasn’t changed for more than a decade. Most cloud integration providers still rely on an ESB architecture and their proprietary connectors. As a result, enterprise integration projects suffer from constraints of availability and reliability of these connectors that are not re-usable across other integration vendors. However, the rapid adoption of APIs and almost ubiquitous availability of APIs amongst most SaaS and Cloud applications are ra...
Jul. 2, 2015 03:18 PM EDT Reads: 715
Agile, which started in the development organization, has gradually expanded into other areas downstream - namely IT and Operations. Teams – then teams of teams – have streamlined processes, improved feedback loops and driven a much faster pace into IT departments which have had profound effects on the entire organization. In his session at DevOps Summit, Anders Wallgren, Chief Technology Officer of Electric Cloud, will discuss how DevOps and Continuous Delivery have emerged to help connect dev...
Jul. 2, 2015 03:00 PM EDT Reads: 1,039
"What Dyn is able to do with our Internet performance and our Internet intelligence is give companies visibility into what is actually going on in that cloud," noted Corey Hamilton, Product Marketing Manager at Dyn, in this SYS-CON.tv interview at 16th Cloud Expo, held June 9-11, 2015, at the Javits Center in New York City.
Jul. 2, 2015 02:52 PM EDT Reads: 696
The Internet of Things is not only adding billions of sensors and billions of terabytes to the Internet. It is also forcing a fundamental change in the way we envision Information Technology. For the first time, more data is being created by devices at the edge of the Internet rather than from centralized systems. What does this mean for today's IT professional? In this Power Panel at @ThingsExpo, moderated by Conference Chair Roger Strukhoff, panelists addressed this very serious issue of pro...
Jul. 2, 2015 02:30 PM EDT Reads: 1,240
Internet of Things is moving from being a hype to a reality. Experts estimate that internet connected cars will grow to 152 million, while over 100 million internet connected wireless light bulbs and lamps will be operational by 2020. These and many other intriguing statistics highlight the importance of Internet powered devices and how market penetration is going to multiply many times over in the next few years.
Jul. 2, 2015 02:15 PM EDT Reads: 2,229
Manufacturing has widely adopted standardized and automated processes to create designs, build them, and maintain them through their life cycle. However, many modern manufacturing systems go beyond mechanized workflows to introduce empowered workers, flexible collaboration, and rapid iteration. Such behaviors also characterize open source software development and are at the heart of DevOps culture, processes, and tooling.
Jul. 2, 2015 02:15 PM EDT Reads: 924
Today air travel is a minefield of delays, hassles and customer disappointment. Airlines struggle to revitalize the experience. GE and M2Mi will demonstrate practical examples of how IoT solutions are helping airlines bring back personalization, reduce trip time and improve reliability. In their session at @ThingsExpo, Shyam Varan Nath, Principal Architect with GE, and Dr. Sarah Cooper, M2Mi’s VP Business Development and Engineering, will explore the IoT cloud-based platform technologies drivi...
Jul. 2, 2015 02:00 PM EDT Reads: 1,090
Containers have changed the mind of IT in DevOps. They enable developers to work with dev, test, stage and production environments identically. Containers provide the right abstraction for microservices and many cloud platforms have integrated them into deployment pipelines. DevOps and Containers together help companies to achieve their business goals faster and more effectively. In his session at DevOps Summit, Ruslan Synytsky, CEO and Co-founder of Jelastic, reviewed the current landscape of...
Jul. 2, 2015 01:45 PM EDT Reads: 2,337
Live Webinar with 451 Research Analyst Peter Christy. Join us on Wednesday July 22, 2015, at 10 am PT / 1 pm ET In a world where users are on the Internet and the applications are in the cloud, how do you maintain your historic SLA with your users? Peter Christy, Research Director, Networks at 451 Research, will discuss this new network paradigm, one in which there is no LAN and no WAN, and discuss what users and network administrators gain and give up when migrating to the agile world of clo...
Jul. 2, 2015 01:36 PM EDT Reads: 696
SYS-CON Events announced today that JFrog, maker of Artifactory, the popular Binary Repository Manager, will exhibit at SYS-CON's @DevOpsSummit Silicon Valley, which will take place on November 3–5, 2015, at the Santa Clara Convention Center in Santa Clara, CA. Based in California, Israel and France, founded by longtime field-experts, JFrog, creator of Artifactory and Bintray, has provided the market with the first Binary Repository solution and a software distribution social platform.
Jul. 2, 2015 01:00 PM EDT Reads: 1,290
"We got started as search consultants. On the services side of the business we have help organizations save time and save money when they hit issues that everyone more or less hits when their data grows," noted Otis Gospodnetić, Founder of Sematext, in this SYS-CON.tv interview at @DevOpsSummit, held June 9-11, 2015, at the Javits Center in New York City.
Jul. 2, 2015 01:00 PM EDT Reads: 1,084
Internet of Things (IoT) will be a hybrid ecosystem of diverse devices and sensors collaborating with operational and enterprise systems to create the next big application. In their session at @ThingsExpo, Bramh Gupta, founder and CEO of robomq.io, and Fred Yatzeck, principal architect leading product development at robomq.io, discussed how choosing the right middleware and integration strategy from the get-go will enable IoT solution developers to adapt and grow with the industry, while at th...
Jul. 2, 2015 12:00 PM EDT Reads: 2,059
Containers are revolutionizing the way we deploy and maintain our infrastructures, but monitoring and troubleshooting in a containerized environment can still be painful and impractical. Understanding even basic resource usage is difficult – let alone tracking network connections or malicious activity. In his session at DevOps Summit, Gianluca Borello, Sr. Software Engineer at Sysdig, will cover the current state of the art for container monitoring and visibility, including pros / cons and liv...
Jul. 2, 2015 11:50 AM EDT Reads: 624
The last decade was about virtual machines, but the next one is about containers. Containers enable a service to run on any host at any time. Traditional tools are starting to show cracks because they were not designed for this level of application portability. Now is the time to look at new ways to deploy and manage applications at scale. In his session at @DevOpsSummit, Brian “Redbeard” Harrington, a principal architect at CoreOS, will examine how CoreOS helps teams run in production. Attende...
Jul. 2, 2015 11:17 AM EDT Reads: 623
"We have a tagline - "Power in the API Economy." What that means is everything that is built in applications and connected applications is done through APIs," explained Roberto Medrano, Executive Vice President at Akana, in this SYS-CON.tv interview at 16th Cloud Expo, held June 9-11, 2015, at the Javits Center in New York City.
Jul. 2, 2015 11:15 AM EDT Reads: 1,089
The cloud has transformed how we think about software quality. Instead of preventing failures, we must focus on automatic recovery from failure. In other words, resilience trumps traditional quality measures. Continuous delivery models further squeeze traditional notions of quality. Remember the venerable project management Iron Triangle? Among time, scope, and cost, you can only fix two or quality will suffer. Only in today's DevOps world, continuous testing, integration, and deployment upend...
Jul. 2, 2015 11:00 AM EDT Reads: 2,170
Malicious agents are moving faster than the speed of business. Even more worrisome, most companies are relying on legacy approaches to security that are no longer capable of meeting current threats. In the modern cloud, threat diversity is rapidly expanding, necessitating more sophisticated security protocols than those used in the past or in desktop environments. Yet companies are falling for cloud security myths that were truths at one time but have evolved out of existence.
Jul. 2, 2015 11:00 AM EDT Reads: 2,234
IT data is typically silo'd by the various tools in place. Unifying all the log, metric and event data in one analytics platform stops finger pointing and provides the end-to-end correlation. Logs, metrics and custom event data can be joined to tell the holistic story of your software and operations. For example, users can correlate code deploys to system performance to application error codes. In his session at DevOps Summit, Michael Demmer, VP of Engineering at Jut, will discuss how this can...
Jul. 2, 2015 10:46 AM EDT Reads: 509
"A lot of the enterprises that have been using our systems for many years are reaching out to the cloud - the public cloud, the private cloud and hybrid," stated Reuven Harrison, CTO and Co-Founder of Tufin, in this SYS-CON.tv interview at 16th Cloud Expo, held June 9-11, 2015, at the Javits Center in New York City.
Jul. 2, 2015 10:45 AM EDT Reads: 776