description Lambda Architecture Overview
Lambda Architecture is a design pattern for processing large data sets by combining a batch layer with a speed or stream-processing layer and a serving layer. The batch path recomputes results from a durable master data set for completeness, while the speed path supplies low-latency updates before those batch results are available. The approach is associated with Nathan Marz's work in the early 2010s and is intended for systems that need both historical accuracy and near-real-time views.
help Lambda Architecture FAQ
What are the three layers of Lambda Architecture?
Lambda Architecture combines a batch layer, a speed or stream-processing layer, and a serving layer. The batch layer recomputes complete results, while the speed layer supplies low-latency updates before the batch results catch up.
Why does Lambda Architecture keep a durable master dataset?
The durable dataset lets the batch layer rebuild views from complete historical data. This helps correct errors or missing events that may have affected the real-time processing path.
What problem does the speed layer solve?
Batch jobs can take too long for applications that need fresh results, so the speed layer processes recent events quickly. A dashboard can therefore show near-real-time information while a later batch job produces the more complete result.
How is Lambda Architecture different from a purely batch system?
A purely batch system may provide accurate results but with significant delay. Lambda Architecture adds stream processing so users can see recent changes without waiting for the next full recomputation.
explore Explore More
Reviews & Comments
Write a Review
Be the first to review
Share your thoughts with the community and help others make better decisions.
