Une infrastructure de stockage et sa suite analytique : Le duo gagnant du Datalake Foundation

Preview:

Citation preview

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

JEUDI 19 NOVEMBRE 2015 Denis FRAVAL-OLIVIER : ISD Presales Manager

DATA LAKE FOUNDATION 2.0

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

Module 4: Horizontal and Vertical Markets

EMC Isilon – Unifying Workloads in one place

© Copyright 2015 EMC Corporation. All rights reserved

Archive & Backup Target

File shares Home Directories

BLOBS

Design, Test & Manufacture

Consumerization Personalization

Splunk

Processes & Transaction

Hadoop & Analytics

Sync ‘n Share

Demographics

Web Content

Social & Next-Gen

Surveillance

ISILON FOR ALL TYPES OF DATA

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

EMC Isilon Scale-Out platform

Clients and Applications

RESTful API GET PUT POST DELETE

Gig-e 10 Gig-e Network

OneFS Operating Environment

Multi-Protocol Client/Application Layer Ethernet Layer

Protocols

SMB NFS

FTP HTTP

HDFS for Hadoop

REST for Object

Intra-cluster Communication

© Copyright 2015 EMC Corporation. All rights reserved

DATA LAKE EMC ISILON SCALE-OUT NAS

DATA PROTECTION

DATA SECURITY PERFORMANCE MANAGEMENT

DATA MANAGEMENT

Data Lake

S-Series X-Series

NL-Series HD-Series

5

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

S - Series X - Series

NL-Series

Isilon CloudPools

3rd Platform cloud Innovation

HD-Series

6

FUTURE

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

NFS

HDFS

SMB, NFS, HTTP, FTP, HDFS

Node reply Node reply Node reply Node reply

Support for Multiple Analytics Applications

name node

name node

name node

name node

data n

od

e

NFS

SMB

SMB

NFS MAP Reduce

MAP Reduce

MAP Reduce

MAP Reduce

MAP Reduce

MAP Reduce

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

Splunk Index Architecture

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

The Big, Cold Data Lake

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

Benefit: Unmatched Scalability = Unmatched Simplicity

Cold Isilon

Single Volume scaling to

50PB

****.gz

****.tsidx

The “Bottomless” Cold Bucket

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

Use Splunk Multi-Site Clustering For Site Protection

Use Array Snapshots For Data Protection

****.gz

****.tsidx

Snapshots

Benefit: Snapshots + Splunk Replication = NO BACKUPS

Hot/Warm XtremIO

****.gz

****.tsidx

Splunk Multi-Site

Clustering

Backupless “Bliss”

Cold Isilon

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

Using Self Encrypting Drives (SED) AES256 Encryption Algorithm

Drives Taken Out Are Unreadable

Benefit: Encryption = Piece Of Mind

Hot/Warm XtremIO

Cold Isilon

Bonus !! Encryption

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

Benefit: Automation = Simplicity

Automated Tiering For High Density Capacity Keep Data In Cold Then Delete

Write-Once-Read-Many (WORM) protection SEC Rule 17a-4(f) definition standards

Bonus !! Always Searchable

Cold Isilon

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

Data-in-place analytics Multi-protocol access

Efficiency gains of 20% vs 300% overhead Enterprise Features For Hadoop

Benefit: Isilon HDFS = SIMPLE Hadoop Analytics

Bonus !! HUNK Ready…

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

Hadoop Market Leadership

Market Leader in Hadoop Shared Storage #1

Customers 700+

YoY Growth 250%

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

Hbase Storm Hive Pig Map Red

YARN

Tez Sqoop Knox Spark Kafka

NameNode

HADOOP ARCHITECTURE - TRADITIONAL

Ethernet

Data Node + Compute Node

Data Node + Compute Node

Data Node + Compute Node

Data Node + Compute Node

Data Node + Compute Node

Data Node + Compute Node

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

Ethernet

DataNode

Compute Node Compute Node Compute Node

Compute Node Compute Node Compute Node

NameNode

name node

name node

name node

name node d

ata no

de

Hbase Storm Hive Pig Map Red

YARN

Tez Sqoop Knox Spark Kafka

Ambari Agent

HADOOP ARCHITECTURE – WITH ISILON

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

Traditional Hadoop - Layers

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

Isilon + Hadoop – NO Layers

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

HDFS

SMB, NFS, HTTP, FTP, HDFS

Node reply Node reply Node reply Node reply

HDFS: Integrated Isilon and vHadoop

name node

name node

name node

name node

data n

od

e

NFS

SMB

SMB

NFS

Apache

© Copyright 2015 EMC Corporation. All rights reserved © Copyright 2015 EMC Corporation. All rights reserved

Si vous avez des questions sur cette présentation, n’hésitez pas à prendre contact directement contact

avec :

Denis FRAVAL-OLIVIER denis.fraval@emc.com

Recommended