Wink is an industry-leading platform that coalesces products from the best brands together into an easy and handy mobile app. Wink allows users to seamlessly monitor, control and automate their home in ways that were never possible before. Their goal is to figure out simpler ways to make a smarter home.
The real challenge for the Wink team was to build a low-latency and highly dependable infrastructure to help millions of connected smart-home devices and consumer hubs to communicate with each other with an added emphasis on horizontal scalability. The prime reason behind seeking horizontal scalability was to be able to encrypt everything quickly so that the connection could easily be revived if ever anything went wrong.
Keeping in view, these technical aspects to ensure a seamless experience for the smart home-connected users, Wink came up with an across-the-board use facilitating Kubernetes-Docker-CoreOS Container Linux stack on a broader scale.
Wink transferred as much as 80% of its workload on a unified stack of Kubernetes-Docker-CoreOS and thus went into a situation to continuously innovate as they have more time to continuously innovate and improve their products and services.
According to Wink’s Head of Engineering Kit Klein, committing to the unified technological stack of Kubernetes-Docker-CoreOS Container Linux smoothed building on the top of the infrastructure, making it a relatively easier task.
Klein had full faith in the adoption of the new technology stack. Post the adoption of this new technology Wink received good commercial reciprocity from two of the biggest American retailers Home Depot and Walmart and it further inspired a lot of confidence in Klein in terms of implementing Kubernetes for the organization. Also, people in general by that time were a fan of Kubernetes because even though many of them didn't have in-depth technical knowledge they all knew this much that they held no space for technical mishaps and always wanted some free time to focus on technical innovation and optimization of its enforcement.
Deploying a Kubernetes-Docker-CoreOS Container Linux stack helped Wink move a step ahead towards its mission statement of making a connected home environment more accessible for end-users.
Building on top of a cloud-native infrastructure backed by Kubernetes-Docker-CoreOS Container Linux stack offered the following advantages to the end-users:
Kubernetes is an open-source technology and it’s really portable.
Different workloads can be distributed across a hybrid cloud or multi-cloud environment.
With Kubernetes, one can always deploy a hybrid cloud on AWS or can even bring it to their homegrown data center.
The stack also provided massive security benefits for those who had one distro/machine image to validate.
Such a stack allows the end-users to save a lot of money and time.
The concept of connected homes was more affordable and reliable for the new users with this technological stack.
With the right infrastructure Wink was able to able to design and produce more smart home products in a prolific way and gained huge popularity in the Home Depot stores & Walmart, the top-notch American stores at that time.
Amazingly, the Wink team deployed the cloud-native infrastructure backed by Kubernetes at the top of CoreOS Container Linux, early in the life of the infrastructure deployment and this helped them overcome the challenge of serving all the communications between hub and products while maximizing the reliability and latency at the same time, right with their first project.
Adopting CoreOS/Kubernetes simultaneously ensured that the Wink team had the much-needed portability in their technological stack which in turn permitted the scope of a lot of plug-ins and integrations.
Deploying a cloud-native infrastructure backed by Kubernetes ensured that many quintessential deployment requirements were taken care of including horizontal scalability, the potential to encrypt details really quick, and auto-healing functionality that ensured that connections or contacts even if lost due to technical glitches could be easily brought back up. The technological stack by Wink in the true sense was a secure socket-based offering for the end-users.
As the Wink team had always used some cluster technology to deploy their technical stack, with the cloud-native infrastructure backed by Kubernetes-Docker-CoreOS Container Linux stack this concept of operationalizing tasks on clusters was intact within a containerized environment that ran in Docker.
Kubernetes helped the team in managing the containerized environment and developing an innate system that helps in distributing the workloads properly across the nodes.
With Kubernetes, the Wink team had the flexibility and portability needed to manage workload across different cloud providers. For example, they could easily run a hybrid AWS or could even bring in their own data center.
Having everything unified on a single Kubernetes-Docker-CoreOS Container Linux stack strengthened the security paradigms and enabled the team to save time and money.
Managing the connected smart-home products for their customers with the help of a unique cloud-native technological stack backed by Kubernetes for the New York City-based company Wink, didn’t come without a fair share of challenges.
Some prominent ones are as under:
They tried a lot many different technological combinations before finalizing a Cloud-Native Infrastructure Backed by a Kubernetes-Docker-CoreOS Container Linux stack. However, as Klein explained the quest also ensured that they discovered a maximum viable product in the process of moving very fast in the process of launching their first product in the marketplace.
A major concern was the side effect of the firewall, whether Wink products would be able to connect to customers' homes. The team had no clue what ports would open behind the firewall and they didn't have anything like a URL. Opening up the real-time, bi-directional communication between the cloud and the user-end devices was the real challenge for Klein and the team as much as it was vital to keep the communication persistent.
The team was focused on depreciating the overhead sending of a message as it would delay the time of service delivery and would adversely impact the end-users' experiences. In that way, minimizing the latency was a core challenge for the Wink Team. The most primitive version of Wink Hub executed simple users' requests like turning on/off the light bulbs by sending these requests to the cloud. Wink software was subsequently updated to enable local control and ensure that the latency period was curtailed down to as low as 10 milliseconds for many devices.
Low-latency internet connectivity still remains an area of concern for the team as the ecosystem of smart-home products is continuously growing, and with this, there arises an ever-increasing need for cloud-enabled integrations.
In 2015, Docker was not prolific in use despite being understood so the Wink team had to start looking out for the potential alternative solutions that existed.
Another limiting factor was the deployment of multi-port non-HTTP/HTTPS services, which was a bit of a daunting task for early cluster technologies because they didn't support port mapping and exposing with ease.
The initial project of Wink therefore despite being very ambitious was too inclined towards HTTP workloads and could only be used on some miscellaneous stuff for a while and could never be in actuality deployed in the production environment.
As simple as it was to decide about a containerized workload, taking decisions about OS and container orchestration platform wasn’t an easy task at all.
They also were a bit worried about having a proper housekeeping infrastructure in place for their ambitious technological stack, for example having a backup plan to restart a container if it accidentally dies with the help of a load balancer. Later with the load balancing and auto-healing capabilities of Kubernetes, these problems found a resolution.
Initially, the Wink team was using Fleet as a basic container orchestration enabler which didn't facilitate routing, sharing configurations, secrets, et cetera, among instances of service. For these additional layers of functionalities to be added one had to write unit files manually which was a bit of a tedious task for the team. So, the team acceded to automating some of these additional facilitations which became possible with the Kubernetes-Docker-CoreOS Container Linux stack
Wink team settled for CoreOS Container Linux because of the container-optimized Linux distribution system that is facilitated. Initially, they were considering building directly on a general-purpose Linux distro like Ubuntu but later they realized that this wouldn't work as they will have to go through multiple additional installations. Additionally, they would also have to avail of a cluster management system like Mesos that could handle larger teams and workloads. Furthermore, CoreOS Container Linux came with a built-in container orchestration system called Fleet and came with an easy-to-use API. Fleet was sufficient for the time being until the team opted Kubernetes for to address some of the heavier workloads.
Klein illustrated that they could trust Kubernetes as it is an open-source community maintained by CNCF which in turn is supported by many experienced people who possess considerable experience in cluster management.
The adoption of Cloud-Native Infrastructure Backed by Kubernetes-Docker-CoreOS Container Linux stack is a great example of how to adapt the cutting-edge technology. It's scary for some people, but if you can stay close to the development and understand why decisions are being made, then it helps you understand how to build your system in harmony with those systems as opposed to trying to work against them.
When it comes to adopting cloud-native infrastructure, Kubernetes, Docker, and CoreOS are the new kids on the block. But before you get too excited about this new technology stack, it's important to understand why it's so popular and how it works.
As cloud-native computing continues to grow in popularity and adoption, there are a number of challenges that organizations face. One of the biggest challenges is understanding what the future holds for IT. The Wink Team set a perfect example of being able to navigate these challenges by adopting Cloud-Native Infrastructure (CNI), backed by Kubernetes-Docker-CoreOS container Linux.
The adoption of Cloud-Native Infrastructure (CNI), backed by the Kubernetes-Docker-CoreOS container Linux stack by Wink can teach major lessons to other organizations. The paramount one is that advancing your organization's IT infrastructure requires a constant commitment to innovation and learning so that you can stay ahead of the curve. That's why it's so important that organizations are able to understand what drives their technology decisions and be able to communicate these decisions clearly and accurately with their stakeholders.
Things have changed a lot since Wink was acquired by Flex in 2015. Now, the company controls 2.3 million connected devices in households all over the country.
Does that mean there's nothing left to do? Not at all! In fact, it's just the opposite: new versions of the hub – Wink Hub 2 – hit shelves – and are being offered for the first time at Walmart stores in addition to Home Depot.
"Two of the biggest American retailers are carrying and promoting the brand and the hardware," Klein says proudly – though he adds that "it really comes with a lot of pressure. It's not a retail situation where you have a lot of tech enthusiasts. These are everyday people who want something that works and have no tolerance for technical excuses." And that's further testament to how much faith Klein has in the infrastructure that this team has built over so many years: they know what they're doing, and they know how to deliver on their promises.
On top of the new hub, Wink has been busy making sure its app is up to date with all the latest features.
"We've been working on the app for a while now," Klein says. "It's really important to us."
But what's next for Wink? Klein thinks that one of the best ways to keep people engaged with their smart home is to keep them entertained. That's why he's been working hard on integrating content from YouTube and Amazon into Wink.
"We're not just selling the hardware," Klein says. "We're here to help our customers be more productive or happier in whatever way they want."
The Wink hub continues to be one of the most popular home automation products out there, and it's easy to see why. The product has been around since 2015, and it allows you to control your lights and thermostat with a few simple taps on your phone.
Wink Hub 2 works with both Amazon Alexa and Google Home, so you can use it with almost any smart home device that you already own. You can also set up scenes that will automatically turn off all of your lights when your kids fall asleep or turn on all of them when they get up in the morning. And if you don't have smart devices around? No problem! The Wink Hub 2 works with any standard light switch or plug-in outlet (no WiFi required). And if anything goes wrong during setup? Just call up customer support—they'll walk you through everything from step one until everything works perfectly!
The engineering team at Wink has grown, and Klein is excited about the machine learning work that's going on behind the scenes.
"We built a system of containerized small sections of the data pipeline that feed each other and can have multiple outputs," he says. "It's like data pipelines as microservices."
Again, Klein points to having a unified stack running on CoreOS Container Linux and Kubernetes as the primary driver for the innovations to come. "You're not reinventing the wheel every time," he says. "You can just get down to work."
The unique thing about Wink's adoption of Cloud-Native Infrastructure Backed by Kubernetes-Docker-CoreOS Container Linux stack was that they were not trying to be a quick-fixes but were focused on building a very solid engineering team who can build on top of our platform.
This focus on reliability has led to some innovations in the company's products. For example, Klein points out how much they've improved their machine learning capabilities by using containers and microservices to build up a unified stack of their data pipeline. "It's like data pipelines as microservices," he says. And he hopes that this approach will allow Wink to continue innovating and improving its products.
We believe that the cloud-native infrastructure approach taken by Wink is a great choice for keeping your smart homes connected. It allows you to deploy a highly scalable platform that is capable of handling the high throughput and low latency requirements for IoT applications.