- This topic has 1 reply, 1 voice, and was last updated 5 years, 7 months ago by
.
-
Posts
-
-
-
Below are few notable features of Spark which make it so powerful :
1. Fast processing of Data- – We can achieve high data processing speed of about 100x faster in memory and 10x faster on the disk.
2. Real-time data processing- The problem with MapReduce was that it can only handle and process data which is already present, but not in the real-time.
3 .Fault-tolerant- It provides fault tolerance through Spark abstraction-RDD. Spark RDDshandles the failure of any worker node in the cluster.
4. Supported by multiple languages – In Spark, there is support for many languages like Java, R, Scala, Python. Thus, it provides dynamicity and overcomes the limitation of Hadoop.
5. Computations in memory as well- The data is being cached so we don’t have to fetch data from the disk every time thus the time is saved.
6.Evaluates lazily- All the transformations in Spark RDDs are lazy in nature. It does not give the result right away rather a new RDD is formed from the existing one.
7. Single processing engine for multiple loads- We can easily develop a parallel application, as Spark provides many high-level operators.
8. Support for Sophisticated Analysis- Spark comes with many tools for streaming data, interactive/declarative queries, machine learning which add-on to map and reduce.
9. Integrated with Hadoop- Spark can run on YARN cluster Manager as well as independently, and thus it can read existing Hadoop data.
10.Cost Efficient- Apache Spark is a cost-effective solution for Big data problem. Because in Hadoop large amount of storage and the data center is required during replication.
To learn more about features visit : Features of Apache Spark
-
- You must be logged in to reply to this topic.
