From the course: AI-Powered Time Series Forecasting with Python
Unlock the full course today
Join today to access over 24,700 courses taught by industry experts.
What are near real-time systems? - Python Tutorial
From the course: AI-Powered Time Series Forecasting with Python
What are near real-time systems?
Near-real-time forecasting, typically just called near-time forecasting, is a powerful approach that can be sufficient for many business needs. First, let's define what we mean by near-time. A near-time system processes and delivers data with a slight delay, typically ranging from a few seconds to a few minutes. It strikes a balance between the immediacy of real-time systems and the batch processing of historical data. That makes near-time systems very suitable for many business cases where a small delay is acceptable. Now let's look at the characteristics of near- time systems. These systems process data at frequent regular intervals such as every few minutes. They provide results or insights with a short delay. And to do this, near-time systems often utilize micro-batch processing techniques that offer a compromise between latency and computational efficiency. So when should you consider near time systems? One scenario is when you need frequent data updates, but you don't…
Contents
-
-
-
-
(Locked)
What are near real-time systems?2m 40s
-
(Locked)
Requirements for near real-time forecasting systems3m 42s
-
(Locked)
Recalculating features4m 41s
-
(Locked)
Frequency considerations2m 26s
-
(Locked)
Online prediction1m 58s
-
(Locked)
End-to-end example2m 23s
-
(Locked)
Advantages and disadvantages of near real-time3m 1s
-
(Locked)
Challenge: Feature Y1m 26s
-
(Locked)
Solution: Feature Y5m 47s
-
(Locked)
-
-
-