Combine ignoring late data with an effective arriv...
# help
j
Combine ignoring late data with an effective arrival time by layering multiple instances of this strategy. My question is : What is multiple instances of this strategy? Set a sequence of deadline intervals. Data goes into the first layer not beyond its deadline, giving a quantized arrival time. My question is : What is quantized arrival time?What quantized means?
a
Hi! "Quantized" here means divided into intervals. For instance, say you use 5-minute intervals. You don't store arrival times on your data points. Instead you just store data into an interval corresponding to its arrival time. So if data arrive at times 12h02m5s, 12h04m24s, 12h05m09s, then the first two points will go into a bucket for points arriving 12h00m -- 12h05m, and the third point will go into a bucket for points arriving 12h05m -- 12h10m. The bucket containing a point gives its arrival time, but only as a 5m-long interval. Does this help?
lakefs 2
j
Thank you!