Streaming Data Dominates: Over 2000 Developers Say “Only Batch” Is Almost Extinct
In our latest survey, Need For Speed: Fast Data Development Trends, over 2400 developers share their use of batch or streaming for data processing workloads The results show that almost 90% of respondents are using some kind of real-time component...
When the Internet’s pioneers were struggling to gain control of their ballooning data sets, the “classic” Apache Hadoop architecture solved the primary use case of batch-mode analytics and data warehousing.
Although batch continues to play a role, the results also reveal that Apache Hadoop may not be relevant for fast data use cases—83% of Fast Data systems today are not running on Apache Hadoop. The results ranked are as follows:
- 33% - Mostly batch, a little real-time
- 27% - Mostly real-time, some batch
- 25% - Equal amounts batch and real-time
- 11% - All batch, no real-time
- 4% - All real-time processing
For these respondents, speed really matters. With 56% using real-time components at least half the time, this shows that enterprises are off to the races and transforming to embrace streaming.
If this data point was interesting to you, then get more insights from over 2400 developers on the impact of “data in motion” in the real world in the full report (PDF):