This document discusses using Alluxio with Spark to improve performance. It begins with an overview of Alluxio and how it fits into the data ecosystem. Common use cases with Spark include accelerating I/O to remote storage, sharing data across jobs at memory speed, and managing data across storage systems. Using Alluxio with Spark consolidates memory, provides data resilience during crashes, and allows easy access to Alluxio data from Spark. Performance evaluations show Alluxio providing 2-17x speedups over Spark for reading RDDs and DataFrames from both local SSD and remote S3 storage.

1. EFFECTIVE SPARK WITH ALLUXIO
Spark Summit EU
Jiri Simsa, Alluxio, Inc.

2. OUTLINE
• Alluxio Overview
• Alluxio + Spark Use Cases
• Using Alluxio with Spark
• Performance Evaluation
2

3. 3
BIG DATA ECOSYSTEM YESTERDAY

4. BIG DATA ECOSYSTEM TODAY
…
…
3

5. …
…
BIG DATA ECOSYSTEM ISSUES
3

6. BIG DATA ECOSYSTEM WITH ALLUXIO
…
…
FUSE Compatible File SystemHadoop Compatible File System Native Key-Value InterfaceNative File System
GlusterFS InterfaceAmazon S3 Interface Swift InterfaceHDFS Interface
3

7. BIG DATA ECOSYSTEM WITH ALLUXIO
…
…
FUSE Compatible File SystemHadoop Compatible File System Native Key-Value InterfaceNative File System
Unifying Data at Memory Speed
GlusterFS InterfaceAmazon S3 Interface Swift InterfaceHDFS Interface
3

8. WHY ALLUXIO
4
Co-located compute and data with memory-speed access to data
Virtualized across different storage systems under a unified namespace
Scale-out architecture
File system API, software only

9. 5
Unification
New workflows across any
data in any storage system
Orders of magnitude
improvement in run time
Choice in compute and
storage – grow each
independently, buy only
what is needed
Performance Flexibility
BENEFITS

10. 6
Popular Open Source Projects’ Growth
Spark Kafka Cassandra HDFS
FASTEST GROWING BIG DATA PROJECTS

11. 6
• Fastest growing
open-source project
in the big data
ecosystem
• 300+ contributors
from 100+
organizations
• Running in large
production clusters
Popular Open Source Projects’ Growth
Alluxio
Spark Kafka Cassandra HDFS
FASTEST GROWING BIG DATA PROJECTS

12. OUTLINE
• Alluxio Overview
• Alluxio + Spark Use Cases
• Using Alluxio with Spark
• Performance Evaluation
7

13. ACCELERATE I/O TO/FROM REMOTE STORAGE
Baidu’s PMs and analysts run
interactive queries to gain insights
into their products and business
• 200+ nodes deployment
• 2+ petabytes of storage
• Mix of memory + HDD
ALLUXIO
Baidu File System
8

14. ACCELERATE I/O TO/FROM REMOTE STORAGE
The performance was amazing. With Spark SQL
alone, it took 100-150 seconds to finish a query;
using Alluxio, where data may hit local or
remote Alluxio nodes, it took 10-15 seconds.
- Baidu
RESULTS
• Data queries are now 30x faster with Alluxio
• Alluxio cluster runs stably, providing over 50TB
of RAM space
• By using Alluxio, batch queries usually lasting
over 15 minutes were transformed into an
interactive query taking less than 30 seconds
Baidu’s PMs and analysts run
interactive queries to gain insights
into their products and business
• 200+ nodes deployment
• 2+ petabytes of storage
• Mix of memory + HDD
ALLUXIO
Baidu File System
8

15. SHARE DATA ACROSS JOBS AT MEMORY SPEED
Barclays uses query and machine
learning to train models for risk
management
• 6 node deployment
• 1TB of storage
• Memory only
ALLUXIO
Relational Database
9

16. SHARE DATA ACROSS JOBS AT MEMORY SPEED
Thanks to Alluxio, we now have the raw data
immediately available at every iteration and
we can skip the costs of loading in terms of time
waiting, network traffic, and RDBMS activity.
- Barclays
RESULTS
• Barclays workflow iteration time decreased
from hours to seconds
• Alluxio enabled workflows that were
impossible before
• By keeping data only in memory, the I/O cost
of loading and storing in Alluxio is now on the
order of seconds
Barclays uses query and machine
learning to train models for risk
management
• 6 node deployment
• 1TB of storage
• Memory only
ALLUXIO
Relational Database
9

17. MANAGE DATA ACROSS STORAGE SYSTEMS
• 200+ nodes deployment
• 6 billion logs (4.5 TB) daily
• Mix of Memory + HDD
ALLUXIO
Qunar uses real-time machine
learning for their website ads.
10

18. MANAGE DATA ACROSS STORAGE SYSTEMS
We’ve been running Alluxio in production for
over 9 months, Alluxio’s unified namespace
enable different applications and frameworks
to easily interact with data from different
storage systems.
- Qunar
RESULTS
• Data sharing among Spark Streaming, Spark
batch and Flink jobs provide efficient data
sharing
• Improved the performance of their system with
15x – 300x speedups
• Tiered storage feature manages storage
resources including memory and HDD
• 200+ nodes deployment
• 6 billion logs (4.5 TB) daily
• Mix of Memory + HDD
ALLUXIO
Qunar uses real-time machine
learning for their website ads.
10

19. OUTLINE
• Alluxio Overview
• Alluxio + Spark Use Cases
• Using Alluxio with Spark
• Performance Evaluation
11

20. CONSOLIDATING MEMORY
12
Storage Engine &
Execution Engine
Same Process
• Two copies of data in memory – double the memory used
• Inter-process Sharing Slowed Down by Network / Disk I/O
Spark Compute
Spark
Storage
block 1
block 3
HDFS / Amazon S3
block 1
block 3
block 2
block 4
Spark Compute
Spark
Storage
block 1
block 3

21. CONSOLIDATING MEMORY
13
Storage Engine &
Execution Engine
Different process
• Half the memory used
• Inter-process Sharing Happens at Memory Speed
Spark Compute
Spark Storage
HDFS / Amazon S3
block 1
block 3
block 2
block 4
HDFS
disk
block 1
block 3
block 2
block 4
Alluxio
block 1
block 3 block 4
Spark Compute
Spark Storage

22. DATA RESILIENCE DURING CRASH
14
Spark Compute
Spark Storage
block 1
block 3
HDFS / Amazon S3
block 1
block 3
block 2
block 4
Storage Engine &
Execution Engine
Same Process

23. DATA RESILIENCE DURING CRASH
15
CRASH
Spark Storage
block 1
block 3
HDFS / Amazon S3
block 1
block 3
block 2
block 4
• Process Crash Requires Network and/or Disk I/O to Re-read
Data
Storage Engine &
Execution Engine
Same Process

24. DATA RESILIENCE DURING CRASH
16
CRASH
HDFS / Amazon S3
block 1
block 3
block 2
block 4
Storage Engine &
Execution Engine
Same Process
• Process Crash Requires Network and/or Disk I/O to Re-read
Data

25. DATA RESILIENCE DURING CRASH
17
Spark Compute
Spark Storage
HDFS / Amazon S3
block 1
block 3
block 2
block 4
HDFS
disk
block 1
block 3
block 2
block 4
Alluxio
block 1
block 3 block 4
Storage Engine &
Execution Engine
Different process

26. DATA RESILIENCE DURING CRASH
18
• Process Crash – Data is Re-read at Memory Speed
HDFS / Amazon S3
block 1
block 3
block 2
block 4
HDFS
disk
block 1
block 3
block 2
block 4
Alluxio
block 1
block 3 block 4
CRASH Storage Engine &
Execution Engine
Different process

27. ACCESSING ALLUXIO DATA FROM SPARK
19
Writing Data Write to an Alluxio file
Reading Data Read from an Alluxio file

28. CODE EXAMPLE FOR SPARK RDDS
20
Writing RDD
to Alluxio
rdd.saveAsTextFile(alluxioPath)
rdd.saveAsObjectFile(alluxioPath)
Reading RDD
from Alluxio
rdd = sc.textFile(alluxioPath)
rdd = sc.objectFile(alluxioPath)

29. CODE EXAMPLE FOR SPARK DATAFRAMES
21
Writing DataFrame
to Alluxio
df.write.parquet(alluxioPath)
Reading DataFrame
from Alluxio
df = sc.read.parquet(alluxioPath)

30. OUTLINE
• Alluxio Overview
• Alluxio + Spark Use Cases
• Using Alluxio with Spark
• Performance Evaluation
22

31. ENVIRONMENT
Spark 2.0.0 + Alluxio 1.2.0
Single r3.2xlarge instance (61GB RAM)
Comparisons:
• Alluxio
• Spark Storage Level: MEMORY_ONLY
• Spark Storage Level: MEMORY_ONLY_SER
• Spark Storage Level: DISK_ONLY
23

32. Reading Cached RDDTime[seconds]
0
75
150
225
300
RDD Size [GB]
0 13 25 38 50
Alluxio (textFile) Alluxio (objectFile) DISK_ONLY
MEMORY_ONLY_SER MEMORY_ONLY
24

33. 25
New Context: Read 50 GB RDD (SSD)
No Alluxio
Alluxio
(objectFile)
Alluxio
(textFile)
Time [seconds]
0 55 110 165 220
2x speedup
4x speedup

34. 26
New Context: Read 50 GB RDD (S3)
No Alluxio
Alluxio
(objectFile)
Alluxio
(textFile)
Time [seconds]
0 200 400 600 800
7x speedup
16x speedup

35. 27
Reading CACHED DATAFRAME (parquet)Time[seconds]
0
62.5
125
187.5
250
DataFrame Size [GB]
0 13 25 38 50
Alluxio (textFile) DISK_ONLY MEMORY_ONLY_SER
MEMORY_ONLY

36. 28
New Context: Read 50 GB DATAFRAME
(SSD)
No Alluxio
Alluxio
Time [seconds]
0 60 120 180 240
2.5x speedup

37. 29
New CONTEXT: Read 50 GB DataFrame
(S3)
No Alluxio
Alluxio
Time [seconds]
0 250 500 750 1000 1250 1500 1750
10x speedup (avg), 17x speedup (peak)

38. CONCLUSION
30
• Easy to use with Spark
• Better Performance
• Predictable Performance

39. Contact: jiri@alluxio.com
Twitter: @jsimsa
Websites: www.alluxio.com and www.alluxio.org
Thank you!
31