From big data to big value : Infrastructure need and Huawei best practise

19
This document is offered compliments of BSP Media Group. www.bspmediagroup.com All rights reserved.

Transcript of From big data to big value : Infrastructure need and Huawei best practise

Page 1: From big data to big value : Infrastructure need and Huawei best practise

This document is offered compliments of BSP Media Group. www.bspmediagroup.com

All rights reserved.

Page 2: From big data to big value : Infrastructure need and Huawei best practise

1 1

HUAWEI TECHNOLOGIES CO., LTD.

FROM BIG DATA TO BIG VALUE

INFRASTRUCTURE NEEDS

AND HUAWEI BEST PRACTICE

HU YUHAI

MARKETING DIRECTOR, BIG DATA & CLOUD STORAGE

HUAWEI ENTERPRISE IT

NOVEMBER 2013

Page 3: From big data to big value : Infrastructure need and Huawei best practise

2

DATA-DRIVEN INSIGHT

Capture Store Process Insight

Making better, more informed decisions, faster

Raw Data

Page 4: From big data to big value : Infrastructure need and Huawei best practise

3

DATA LANDSCAPE CONTINUES TO EVOLVE

HUMAN Generated

UNSTRUCTURED DATA

Email

Web Logs

Documents

MACHINE Generated

SEMI-STRUCTURED DATA

Satellite

Images

Bio-

Informatics

M2m Log

Files

Sensors

Video

BUSINESS PROCESS

Generated

STRUCTURED DATA

OLTP

1990 2000 2008 2013

Data Volume Captured and processed

Data Velocity Of ingest and time

sensitivity for analysis

Data Variability Data format

Audio

Social

Page 5: From big data to big value : Infrastructure need and Huawei best practise

4

BIG DATA ANALYTICS DATA FLOWS

MPP DW

ERP

SCM

CRM

OLT

P

Capture Store Process Insight

ETL OLTP DB

Terabytes

MPP Data Store

Converged Compute & Storage

Machin

e

Exabytes

SAN

Web Logs

NAS

Hum

an

Petabytes

Page 6: From big data to big value : Infrastructure need and Huawei best practise

5

EXAMPLE FOR “EXABYTE” REQUIREMENT

"CERN is hitting the technology limits for resource-intensive simulations

and analysis. Our collaboration with Huawei shows an exciting new

approach, where their novel architecture extends the capabilities in

preparation for the Exascale data rates and volumes we expect in the

future." said Bob Jones, head of CERN OpenLAB

Page 7: From big data to big value : Infrastructure need and Huawei best practise

6

INFRASTRUCTURE REQUIREMENTS

Scale capacity on demand

Scale bandwidth on demand

High throughput ingest

Process data in place near real-time

Cost effective, follows Moore’s Law

EXISTING INFRASTRUCTURE DOESN’T SCALE !

Scaling in every dimension is key !

Page 8: From big data to big value : Infrastructure need and Huawei best practise

7

INFRASTRUCTURE NEEDS

Scale-out distributed storage platforms

‒ Bring the computation to the data

‒ Can’t move Petabytes around network

‒ High throughput streaming workloads

‒ Batch oriented processing

Colum-oriented NOSQL and MPP databases

‒ Flexible schemas, massive scale

Real time analytics requires massive flows

‒ New platforms combine real-time with batch

‒ Trigger on events and process historical data

Page 9: From big data to big value : Infrastructure need and Huawei best practise

8

Huawei Strategy

“Build the Most Efficient Big Data Platform”

Infrastructure is Key of Big Data

Scale out and X86 architecture, all IP based

Fully symmetric and distributed file system

Intelligent Application Awareness

Multi protocol Interface

Openness and cooperation

Natively support Multi-workload

Integrated Storage, analysis and archiving functions

Data full life cycle management

HUAWEI STRATEGY ON BIG DATA

Page 10: From big data to big value : Infrastructure need and Huawei best practise

9

“HIGH SCALABILITY” DISTRIBUTED STORAGE SYSTEM

NFS/CIFS/HDFS

High Performance

Store and Archive

Query and Retrieval

for Structured Data

SQL

Analysis Processing

for Unstructured Data WORKLOAD

STANDARD EXPOSURE

HTTP/S3

EB-level Storage

Resource Pool Mgmt

OCEANSTOR BIG DATA

FRAMEWORK

MR/HBASE

DISTRIBUTED

RAID

LOAD

BALANCE

QUOTA

MGMT

STORAGE

TIERING

TELECOM M&E BANKING GOVERMENT ENERGY

MPP DB

ENGINE

ENTERPRISE HADOOP

ENGINE

OBJECT STORAGE

ENGINE

NATIVE

INTERFACE NATIVE

INTERFACE HDFS

• World Leading Performance and Scalability Storage

Platform as the Infrastructure.

• Natively Integrated HADOOP, MPP DB, OBJECT

Engine, Efficient Data Loading and Processing.

• End-To-End Data Protection and Life Cycle Mgmt.

HUAWEI ENTERPRISE-LEVEL BIG DATA PLATFORM

Page 11: From big data to big value : Infrastructure need and Huawei best practise

10

No.1 Scalability

288 Nodes

No.1 Capacity

40 PB

No.1 Performance

5,000,000 OPS

OCEANSTOR 9000 BIG DATA STORAGE

Performance 3x 1,112,705 1,512,784 1,564,404

3,064,602

5,000,000

500,000

1,000,000

1,500,000

2,000,000

2,500,000

3,000,000

3,500,000

4,000,000

4,500,000

5,000,000

5,500,000

EMC Isilon NetAppFAS6240

AvereFXT3500

OceanStorN8000

OceanStor9000

5,000,000 OPS

Page 12: From big data to big value : Infrastructure need and Huawei best practise

11

Customized Hadoop

‒ Reliability improvements

‒ Redundancy, Failover, SPoF elimination

‒ Security/privacy improvements

‒ Encryption of data and metadata, KERBEROS access control

‒ Management simplification

‒ GUI platform management tools, role-based admin

‒ All Hadoop tools, such as HIVE, PIG, etc.

Innovative DR Solution

‒ DR site up to 1000km

Special VM instances for Hadoop processing

ENTERPRISE-LEVEL HADOOP PLATFORM

Page 13: From big data to big value : Infrastructure need and Huawei best practise

12

Dashboard – Overall System Status Service Management

Resource Management

MANAGER SNAPSHOT

Page 14: From big data to big value : Infrastructure need and Huawei best practise

13

DISTRIBUTED STORAGE SYSTEM

NFS/CIFS/HDFS SQL HTTP/S3 MR/HBASE

MPP DB

ENGINE

HADOOP

ENGINE

OBJECT

ENGINE

NATIVE

INTERFACE NATIVE

INTERFACE

NATIVE

HDFS

• Multi-Workload Scale-Out

Storage Platform

• Leading Storage Efficiency and

Scalability

• End-To-End Data Protection

• Enterprise-Level Hadoop Model

• Native Integrated Hadoop/

MPP-DB/Object

• Unified Management

DISTRIBUTED

RAID

LOAD

BALANCE

QUOTA

MGMT

STORAGE

TIERING

“OCEANSTOR” BIG DATA PLATFORM HIGH LIGHTS

Page 15: From big data to big value : Infrastructure need and Huawei best practise

14

HVS85T / HVS88T

• Smart Matrix Architecture

• Industry-leading RPO

• RAID2.0+ improves efficiency by

300%

16 Controller

High scalability

1,000,000 IOPS

No.1 performance

7 PB

No.1 capacity

3 TB

No.1 cache

Critical

Business

Centralized

Storage Virtualization DR

HVS: NO.1 PERFORMANCE ENTERPRISE STORAGE

Page 16: From big data to big value : Infrastructure need and Huawei best practise

15

2.1 PB

Capacity / rack

60% Energy saving

45% Reduced TCO

40 GB

Output BW / rack

Universal Distributed Storage

• Native object storage, decentralized architecture

• Unlimited Scalability: EB-level capacity

• Extreme Reliability: 99.9999% data durability

• Low TCO: Energy saving HW & Zero-Touch design

Space Lease

Centralized Backup

Web Disk

Active Archive

UDS: EB-LEVEL MASSIVE STORAGE SYSTEM

ARM based high density hardware

DHT: Highly Available Key Value Store

DHT ring

SoD client Hash (key) P0

P1

P2

P3

P4

P5 P6 P7

P8

P9

P10

Pm

1282/0

Page 17: From big data to big value : Infrastructure need and Huawei best practise

16

Distributed Cloud Data Center

Data Center Facilities

Servers Storage

Cloud

Computing

Converg

ed

Infra

stru

ctu

re

Applications

Mana

gem

ent Big

Data

HUAWEI IT BUSINESS COVERAGE

Page 18: From big data to big value : Infrastructure need and Huawei best practise

17

Big data is here

Big data presents new challenges to infrastructure

Be careful with an open source Hadoop

Implementing a robust foundation and careful selection

of tools can allow you to benefit from big data

KEEP YOUR COMPETITIVE ADVANTAGE

Page 19: From big data to big value : Infrastructure need and Huawei best practise

18

HUAWEI TECHNOLOGIES CO., LTD.

THANKS!