Alluxio: The missing piece of on-demand clusters at Alluxio Meetup 2016

The Missing Piece of On-Demand Clusters

September 2016 @ Samsung Calvin Jia

About Me

•  Calvin Jia

•  Software Engineer @ Alluxio, Inc.

•  Alluxio PMC

•  #1 Alluxio Contributor

•  Twitter: @JiaCalvin

2

Alluxio Inc.

•  Founded by Alluxio creators and top committers •  Formerly Tachyon Nexus, Inc. •  $7.5 million Series A by Andreessen Horowitz •  Committed to the Alluxio Open Source Project •  Company Website: http://www.alluxio.com •  We are hiring!

3  

Outline

•  Cloud Architectures •  Alluxio in Cloud Architectures •  Takeaways

4  

Cloud Architectures – Overview

•  Mostly service based, from providers –  Amazon Web Services –  Google Cloud Platform

•  Separate compute and storage clusters •  Compute clusters are ephemeral

5  

Cloud Architectures – Pros & Cons

Pros •  Low maintenance •  Pay as you go •  Elastic and scalable •  Cost effective storage

Cons •  Lower Performance

6  

Outline

•  Cloud Architectures •  Alluxio in Cloud Architectures •  Takeaways

7  

Alluxio in Cloud Architectures – Overview

•  Deployed in compute clusters –  Memory speed data access –  Transparent data access to any storage

•  Simple to deploy –  Mount storage systems similar to local disks

8  

Alluxio in Cloud Architectures – Example

9  

Alluxio in Cloud Architectures – Benefits

10  

•  Remedies the performance drawback •  Acceleration due to memory-speed I/O •  Designed to improve the affinity of compute

and storage

Alluxio in Cloud Architectures – Data Path

11  

FAST 104 - 105 MB/s

MODERATE 103 - 104 MB/s

SLOW 102 - 103 MB/s

Only when necessary Limited

Often

SSD  HDD  

Mem  

Takeaways – Experiment Results

12  

0  

100  

200  

300  

400  

500  

600  

Ini/al  Read   Subsequent  Read   Read  from  Separate  Job  

Run+

me  (secon

ds)  

Spark  -­‐  No  Persist  

Spark  -­‐  Persist  

Alluxio  

Takeaways – Alluxio & Cloud Architectures

13  

•  Cloud architectures have significant upsides •  Alluxio alleviates the major downsides

Try out Alluxio 1.2.0 http://www.alluxio.org/releases

14  

Read the blog at http://www.alluxio.com/2016/08/accelerating-on-demand-data-analytics-with-alluxio

Read the whitepaper at www.alluxio.com.

Check out www.alluxio.org.

Thank you!

15