In my previous blog post, I introduced the notion of the Visible Internet (today’s app economy), and the Invisible Internet (the structural and management patterns that make the visible Internet possible).
The Internet has revolutionized and transformed business, commerce, and culture and the changes in the Invisible Internet to support all this are no less revolutionary. Let’s take a closer look at the drivers of transformation.
A short history of systems' complexity
In the 1990s, the “invisible” complexity of commonly used software applications began to rapidly outstrip the complexity of common “application” hardware, or tangible assembled physical structures. For example, a 747 plane assembled in the mid 90s has over a million moving physical parts. Meanwhile, Microsoft’s Excel application from the same era has over a million lines of code. However, if you look at the particular ways in which each of those lines of code can interact with another, Excel as a feat of engineering is geometrically more complex than the 747….though that was a formidable accomplishment in and of itself.
Beyond applications themselves, the operating system software that supported the applications was becoming more complicated, as were the microprocessors upon which the operating systems ran. This would be well known by any engineer at Intel faced with helping the company keep up Moore’s Law, or any programmer at Microsoft or Apple working on Windows NT, OSX, or similar recent operating systems.
As the Internet became increasingly popular in the late 1990s and early 2000s, another area of computing – WAN Switching Software - came under intense pressure to evolve rapidly, and support geometrically more complex uses. The visible Internet of this era – characterized by the popular portal sites, and the first popular ecommerce sites - was enabled under the waterline by physical devices such as switches, routers, and load balancers from vendors such as Cisco, running the Internet by routing and switching its traffic and data to web sites comprised of web servers, application servers, and databases.
Traits of the 'turn-of-the-century' switching systems
- High speed: Speed is critical for switching software. Compare with applications and programming tools that can afford to be less performant, or where speed is judged in half second increments rather than tens of milliseconds.
- Complexity: Though they had to be fast, switching software in the 1990s didn’t have to make all THAT many decisions . . .
- Low tolerance for error: Compare to application and tools software, which have a higher tolerance for error.
- Relatively low change frequency: Once a switching system was in place and running, “don’t touch it” became the operative management principle. The operating software would go through a major revision once every year or two - but that didn’t mean that the new version would actually go into production.
- Hosted on “contained hardware”: Meaning single user/tenancy. These applications can take advantage of all the available compute resources.
- Simple disaster recovery models: In its simplest form, the restoration of service of a switching system, if it ever failed, could take the form of simply restarting the device, and perhaps slapping in a routing table. But there was no lengthy data restoration process, and the restart interval typically would be measured in minutes.
Fast Forward to 2013
Cisco switches ran the Internet of ten years ago. But the demands on and requirements of backplane routing and architecture have changed.
Device and IP-based routing is replaced by cloud-based routing.
Social chemistry, business models, data, and organizational structures
Social chemistry, business models, and organizational structures have changed totally in the first decade of this century. The things that people and companies are doing with the Internet today are qualitatively different than they were even 10 to 15 years ago. The worldwide Web has exploded with greater than 1 billion users who generate thousands of petabytes of data monthly. All of it - the app economy, the “Internet of Things”, the explosion of mobile devices and social networking - puts new demands on the backplane.
Companies and their customers and partners (even with employees) are interacting in new ways. Leading companies are creating and using mobile apps for new revenue opportunity; APIs are powering the development of powerful mobile and Web apps, API traffic represents revenue opportunities and growth, and large volumes of data from new apps, APIs, and a myriad other new and traditional sources is the currency of the new economy.
A discussion of the changes in the past decade that has revolutionized the way the Internet works above and below the waterline would not be complete without a discussion of Big Data and Analytics. Together they play a critical role in managing and improving businesses.
There’s also a fundamental shift in the qualitative nature of today’s data. Gathering, evaluating, analyzing, and operationalizing big data from your internal systems-of-record as well as from diverse new sources outside of the walls of the enterprise becomes critical for IT operations as well as for predictions about your business.
Systems and teams produce both content and transactional APIs. Analytics about which APIs are being called the most, about the apps that are being downloaded and used (and why), about the top developers adopting your APIs, and so on, have become the key performance indicators of business models. With the increased access to new data sources and with customer and employee interactions happening on social and business networks, enriching the traditional interactions in enterprise systems of record, companies will pivot and evolve based on new insights gleaned from structured and unstructured data across their entire ecosystem.
Inside the enterprise, an entirely new kind of organizational structure is becoming commonplace. It is not unusual to see the severing of lines of business from corporate IT. Virtual teams with diverse skillsets – from developers to business people to consulting agencies - are formed with the charter and budget to find and implement new business models and revenue streams predicated upon mobile computing and the app economy. They are challenged with doing so more quickly and less expensively than traditional IT processes and budgets will allow, but ironically often do not have strong core technical skills within the group.
Guess what? The Internet backplane model of necessity is changing to support these new usage and business models.
Next time, we’ll take a closer look at the changes in the Internet backplane.