To the edge and beyond!

Edge Computing can increase performance, enhance security, cut costs and enable more reliability / scalability. If you haven’t already embraced this new “last mile” delivery, you may soon.

Amazon released their first public cloud computing service (S3) 15 years ago. Companies, particularly the enterprise, are still migrating to the cloud. In the past three years there has been a shift from cloud computing to Edge Computing.

Edge Computing refers to the computations that take place at the edge of the Internet, as opposed to Cloud Computing, where computations happen at a central place. This “edge” can be a branch / retail location, your CDN provider, your network provider and even your security vendor. Edge Computing can process and analyze data in real-time. This high performance approach opens the door on how data will be processed going forward.

Enterprises are used to computing and storing data using a traditional data center or doing the same thing in the cloud. We can now shift the compute function and cost from the data center or cloud to the edge.

I bet compute is the most expensive line item on your cloud provider bill

Edge Computing is not a replacement for cloud computing. It works to optimize the workflow by transferring compute, cutting down processing time, forwarding the traffic you want and providing data redundancies.

Many industries have already started to adopt Edge Computing including autonomous cars, drones, physical security, IoT, home automation, AR/VR systems, medical devices and more.

How can other industries benefit? Here are some ideas:

  • A/B Testing = offer alternate content to a subset of users to ensure accuracy prior to going live with your new release/feature
  • Geolocation = Target ads to certain areas. Find the closest store location.
  • Autocomplete = cache and store common search terms eliminating call backs to your origin
  • Analytics = inject analytics tags for obeservability
  • URL Decoding = security and uniformity
  • Security List = deny traffic based on request-header, cookie content and more. Check for or strip cookies.
  • Dynamic Content = show personalized and individual content
  • Promotional Campaigns = show a campaign based on returning customer information, geolocation, and more
  • Authentication = JWT tokens and SSO integrations
  • Monitoring = inject Javascript to send metrics to your APM and other monitoring tools
  • User Experience = based on internet connection speed offer video quality based on the best fidelity setting
  • Privacy = GDPR and other regulations, check for consent and make decisions to write the cookie
  • Payment Processing = send number to processor, get back hash/token eliminating the need for storing sensitive data
  • Language Translation = translate your content based on location or preference
  • Signage = change advertisement & offer based on time, location and attendance
  • Cashier-less stores = all data and payment processing is down locally
  • Serverless Compute = lightweight serverless compute (LAMBDA functions) can be moved to the edge
  • Log manipulation = save space and bandwidth to your data center / cloud log storage

Why?

Speed/Performance/Reliability

Traditionally all data is sent to your cloud infrastructure. What if you need to respond faster? What if you have unstable applications / connections? Response time is critical so interpreting the input data close to the source is the best option. If we can do both the input and output behavior at the same location, we remove the network latency and real-time becomes possible. Edge Computing offers massive speed when compared to traditional & cloud computing. Reduced network latency is the key.

Bandwidth

Data is processed and analyzed at the same place without uploading/downloading this data elsewhere. Less data transmission as we can do all of the heavy lifting at the Edge. Instead of sending raw data to the cloud, we can process at the edge. We can also enable a stable data flow when there is intermittent connectivity.

Security

Edge Computing utilizes a distributed architecture and is less vulnerable to DDoS attacks. This distributed nature helps maintain the overall availability. Think of it as having your cloud computing infrastructure across all available regions from your cloud provider. We can also make decisions on which traffic to forward and hand off to our cloud infrastructure.

Scale/Flexibility

More systems means more bandwidth and more load on centralized compute instances. Infrastructure needs cannot be capped if you want to deliver value to your customers. With cloud we have to scale accordingly and linearly to the load. With Edge Computing, we do not need to increase bandwidth and compute cost per request. Another benefit is that we can shut down specific modules instead of an entire system. Whether that is planned or unplanned.

Cost

Edge computing is cheaper when compared to cloud computing. Storage is also cheaper. Edge Computing can present a clear picture of your solutions ability and future projections for better strategy and expansion. Different levels of processing are needed for different categories of data. Edge Computing enables a cost-saving structure by ensuring that only necessary resources are allocated for processing and analysis.

The applications of the future will evolve to demand better performance and more innovation. Everything is interconnected and growing demand for real-time applications and the cost pressures are pushing us to innovate, yet again.

Originally published on LinkedIn
https://www.linkedin.com/pulse/edge-beyond-dale-frohman

Principal Site Reliability Engineer. Cyber Security Professional. Technologist. Leader.