Industry executives and experts share their predictions for 2019. Read them in this 11th annual VMblog.com series exclusive.
Contributed by Marc Verstaen, co-founder and CEO of Anaxi
Moving the Boundaries of the Cloud, in Every Direction
I've been lucky enough to participate in two
explosive growth experiences -- Docker and Apple. In both cases, their products
broke new ground and great success was expected from the very beginning. But
the exponential growth of Docker adoption, and the incredible success of the iOS
platform shocked me. I'm definitely better at predicting the past than the
future. My point is that predictions are hard and inaccurate, but they are fun,
too, so here goes.
I'll cover four different areas, all of
massive interest to me, as a software developer: infrastructure software,
hardware as in processor, networking and developer platform.
#1 I expect to see the takeoff of
serverless, the cloud-computing execution model in which the cloud provider
acts as the server, dynamically managing the allocation of machine resources.
It's already happening, and I expect to see much more in the year ahead. Why?
Because it fills a dire need for agile development, it's such an easy
transition to adopt and pricing is based on the amount of resources consumed by
an application, so it's highly efficient, as well.
All the major cloud providers offer
serverless. AWS was the first with Lambda,
introduced in 2014. Microsoft Azure has Azure Functions and Nuclio, Google
Cloud Platform offers Google
Cloud Functions and IBM has OpenWhisk. Oracle provides Fn
Project on Oracle Cloud Platform, which is available on GitHub for deployment
on other platforms. The main risk factor here is fragmentation: there is a need
for a standard that would allow for picking the cloud platform on the fly. It's
in the best interest of the customer, but I would expect there will likely be
some resistance to this. Don't expect the world to move to serverless
overnight. Existing apps will be modernized using Docker, and the majority of
new projects will still rely on containers. But it's the start of a new trend.
#2 That leads me to my second point on the increased attention to quantum computing. It's very early days in
this field but lots of money and resources are going into research in this
area. A Toronto-based company, D-Wave Systems,
can sell you for a mere $15 million, a quantum computer with about a thousand qbits or IBM already offers cloud access.
What are the capabilities of quantum
computing? It can handle exponential algorithms whose execution times scale
exponentially with the size of the problem. For example, breaking encryption
keys. Breaking two bits key is almost trivial, but breaking a 256 bits key can
take a hundred years. Not for a quantum computer, because the execution time
scales linearly. That means current encryption keys will become useless as
protection for our data, bitcoins and others. It will take something like 10
years for quantum computing to reach mass availability, but it will take time
to enhance our encryption systems so we better pay attention now.
#3 Not so far ahead is quantum
networking. With so much data being generated and so much of it traversing
networks, I believe it is inevitable that we overhaul networking technology as
it's known today. I won't even try to dig into the intricacies of the
underlying science, but once it is widely available, consequences will be
gigantic: instant network connections. No latency. Even the Einsteinian speed
of light limit will be gone. All computers can be considered as one system,
regardless of where they are. Computer science as we know it will have to be
completely revisited. Science fiction? Not really, there are at least two
networks that I know of based on these principles -- one in the Netherlands,
the other one in China. Today, their
"limited" use case is around encryption key, the quantic networks are
used to exchange quantic keys in a secure way. This article talks about the work that
is being done.
I believe with quantum networking there is
tremendous potential to change how we work today. This is especially important
with the transition we're seeing to microservices and cloud-native applications
that have smaller software functions and increased reliance on communications.
Networks, as we know them today, cease to exist. Communication must be
instantaneous.
#4 Now, I'll take a much more practical look at something that today we
think is important for software engineering and that is, a real platform for software collaboration.
Today, we have two sets of tools, with two leaders: GitHub and Jira. They are
both good tools helping manage the process of software development, but each
works separately from the other. Jira can help you manage your tickets, prepare
the flow of work. GitHub will manage your code, and provide a pull request flow
to discuss submissions, one by one. Having two separate tools works well for
pure software companies, which are focused on a limited set of projects that
they control. But it starts breaking up for the rest of the world. Now, every
large company is also a software company. They don't sell software, they
consume what the Open Source world delivers, they create the products they need
to communicate with their customers, produce their products, exchange with
their partners. This results in a large number of interdependent projects, very
often built on top of open source projects. And there are no tools to manage
such a complex world. Teams are constrained in silos.
By aggregating data from different tools, we
can give engineers and managers access to metrics that provide a holistic view
of the development process - not piece parts. That increased visibility and
intelligence on their software development projects and processes helps keep
development schedules on track while connecting them to open source projects
they have little control on. This is an easy prediction because it is something
we are working on delivering early in
2019.
I look forward to re-visiting at the end of
the year to check on progress made in these four areas. We are definitely
living at a very exciting time!
##
About the Author
Marc
Verstaen is the co-founder and CEO of Anaxi, whose
mission is to fix software project management. He has been managing engineering
teams for more than 20 years, lastly at Docker as EVP Product Development. He
co-founded two other startups and was the CEO at TextMe, Beatware and Lorienne.
He also acted as a Senior Manager at Apple, overseeing software development on
iCloud clients, UIKit and iOS development tools, and at Oracle (mobile, UX).