Hadoop vs Spark
The cluster centers on debates comparing a discussed tool or approach to Hadoop, frequently questioning its advantages over Hadoop and positioning Spark as a faster, memory-based replacement for MapReduce while arguing Hadoop is often overkill or outdated.
📉 Falling 0.5x DevOps & Infrastructure
4,254
Comments
20
Years Active
5
Top Authors
#6587
Topic ID
Activity Over Time
2007 8
2008 71
2009 175
2010 176
2011 256
2012 255
2013 320
2014 325
2015 417
2016 393
2017 287
2018 236
2019 319
2020 218
2021 244
2022 123
2023 135
2024 169
2025 103
2026 24
Top Contributors
Keywords
MapReduce HDFS AI CPU SQL HADOOP HIVE ML ODBC JDBC hadoop spark big data data sql clusters hive suits processing data sets
Sample Comments
How is this better than Hadoop?
How is this better than Hadoop?
Why is this better than Hadoop?
Spark is a good replacement for MapReduce. MapReduce != Hadoop.
Because he's using Hadoop :P
What do you see as being better than hadoop?
Don't use Hadoop even if your data is big. Spark replaced it.
Guess that's impossible without Spark or Hadoop.
Obligatory "you don't need Hadoop": https://news.ycombinator.com/item?id=14401399
Seems neat, but why is this better than Hadoop?