Innovation Driving Opportunity for the Web Infrastructure John - - PowerPoint PPT Presentation

innovation driving opportunity for the web infrastructure
SMART_READER_LITE
LIVE PREVIEW

Innovation Driving Opportunity for the Web Infrastructure John - - PowerPoint PPT Presentation

Innovation Driving Opportunity for the Web Infrastructure John Fowler Executive Vice President, Systems Sun Microsystems, Inc. Communications Arrives Smart Mobs Co-Creation Wikis Blogosphere Social The Tags Networks Long Tail Next


slide-1
SLIDE 1

Innovation Driving Opportunity for the Web Infrastructure

John Fowler Executive Vice President, Systems Sun Microsystems, Inc.

slide-2
SLIDE 2

Blogosphere Social Networks Smart Mobs Wikis Co-Creation Tags The Long Tail

Communications Arrives

slide-3
SLIDE 3

Next Generation

Software

slide-4
SLIDE 4
  • Many cores
  • Many threads
  • Open platforms
  • $/Performance
  • $/Gigabyte

OPEN Storage Compute Networking

  • Huge bandwidth
  • Open platforms

This is not your Daddy's

Infrastructure

NOW NOW

COMING

Everything connected

Open standards everywhere Horizontal scaling by definition

slide-5
SLIDE 5

Web 2.0

Architecture

NETWORK STORAGE COMPUTE

External Resources Clients

(load balancing, filtering)

Distributed Object Stores NFS

Storage Nodes

Horizontal Scaling, network intensive Compute Nodes

Database Nodes

(unstructured data)

Web/App Servers Cache Servers Reverse Proxies INTERNET

slide-6
SLIDE 6

STORAGE COMPUTE Web 20kit

Structure

Geocoder Driver

NETWORK

Web/App Server Storage Database

Local FS Mogile

SYSTEMS UNDER TEST

slide-7
SLIDE 7

It’s Driving You

CRAZY!

Performance Power, Heat, Space Understanding the Infrastructure Scale

slide-8
SLIDE 8

Compute

More threads, greater efficiency, lower energy consumption, less heat

Growth of thread-rich applications Higher frequencies = more heat Memory latency compute time

Compute Memory latency

slide-9
SLIDE 9

Cores and Threads Are on the

MOVE

8

CORE

16

CORE

slide-10
SLIDE 10

Application Performance and Memory

Capacity

Memory Capacity Throughput

4GB 16GB 64GB 256GB 1TB

slide-11
SLIDE 11

Open Storage

Servers + Storage + OpenSolaris

  • Built on OpenSolaris
  • Performance of ZFS + SSDs
  • Cost efficiency of volume

Hardware

  • Scale Easily

> Millions of files > Gigabytes/sec

  • Management simplicity

> Analytics with dTrace > Diagnostics with FMA

slide-12
SLIDE 12

Why Applications Don’t Perform

Waiting for DATA

  • Today’s Multi-Core,

Multi-Socket application server designs are increasingly held back by slow storage

  • When requesting data, the

server spends most of its time waiting for storage

  • Application performance

remains sluggish regardless of the server CPU horsepower

  • The traditional remedy of

adding more expensive DRAM may no longer suffice as data sets double every 2 years

Multi-Core, Multi Socket CPUs High Performance DRAM

Server Storage

15K RPM Disk Drives

260 Times Slower

slide-13
SLIDE 13

1

Integrating Solaris (TM) ZFS & Co- Developing with Industry Leaders

E n t e r p r i s e S S D

  • u

p t

  • 5

,

  • 8

, W r i t e I O P S

  • u

p t

  • 3

,

  • 4

, R e a d I O P S

  • 3

2 G B $ p e r I O P S : . 8 E n t e r p r i s e H D D

180 Write IOPS 320 Read IOPS 146GB

$ p e r I O P S : 2 . 4 3

NEWS TODAY!

COMING SOON!

slide-14
SLIDE 14

Reliability Comparisons

Bad Block Management

  • Wear leveling
  • ECC Correction
  • No Moving Parts

!

  • Power Consumption

13 Watts 2 Watts Thermal Operating Range 5-55 Celsius 0-70 Celsius

MTBF

1.2 Million 2 Million

slide-15
SLIDE 15

ZFS* Hybrid Storage Pool Model

Automated Data Management

ZIL SSDs Read/ Write Cache SSDs

*Other names and brands may be claimed as the property of others.

High Performance Read & Write Cache Pool

ZFS combines main memory and SSDs for

read caching

Adaptive Replacement Cache (ARC)

ZFS Intent Log pool

ZFS is a transactional file system Intent Log stores small transactions without

doing the transaction

ZFS can use a separate intent log device

(slog)

Typically NVRAM or dedicated SDD pool used

High Capacity Pool

The ZFS pool manager automatically stripes

across any number of high capacity HDDs

ZFS IO-scheduler bundles disk IO to optimize

arm movement and sector allocation Shipping Now in

HDD Pool

slide-16
SLIDE 16

New Server Memory Hierarchy

L1 Cache L2 Cache L3 Cache RAM FLASH Disk

Pico Sec. Nano Sec. Micro Sec. Milli Sec.

slide-17
SLIDE 17

1 1

The Hybrid Storage Pool

PCI-e 8x 6x SAS 3G 1066MT FSB 1066MT FSB 1066MT FSB 1066MT FSB DDR2 667 DDR2 667 DDR2 667 DDR2 667

Xeon 7350 2.93Mhz (16 cores) Quad Channel FB-DIMM 32 DIMM Slots

845,000 IOPS 840,000 IOPS

840,000 IOPS

Performance 24x SSD Capacity Nx 1TB HDD

Sun estimation of performance with assumption of 18,000 IOPs per Ghz per Core for System. 200 IOPS per HDD and 35,000 IOPS per SSD

Gaining Performance and Capacity Balance

slide-18
SLIDE 18

1 1

Traditional Storage Pool Hybrid Storage Pool

SunFire X4450 Server 4 Xeon 7350 Processors (16 cores) 32GB FB DDR2 ECC DRAM

(5) 400GB 4200RPM SATA Drives (7) 146GB 10,000RPM SAS Drives (1) 32GB SSD ZIL Device (1) 80GB SSD Cache Device

Hybrid Storage Pool in Action

Up to 3x the Performance and Reducing Power Consumption by 5x

Before 2010, Flash will be in Every Server

slide-19
SLIDE 19

Sun Studio ZFS/ Solaris

Our

Communities

slide-20
SLIDE 20

Web 2.0

Architecture

NETWORK STORAGE COMPUTE

External Resources Clients

(load balancing, filtering)

Distributed Object Stores NFS

Storage Nodes

Horizontal Scaling, network intensive Compute Nodes

Database Nodes

(unstructured data)

Web/App Servers Cache Servers Reverse Proxies INTERNET

slide-21
SLIDE 21

Let's Break the Rules

Together

  • Visit the Sun booth today

> See our demos > Attend our presentation on the

Web20kit at 3:45PM, Salon A-D

  • Learn more about working with us

> Try and Buy > Start Up Essentials

  • Join the Sun Developer Network

> developers.sun.com/services

  • Test drive and download our

technologies

Visit Us sun.com/web

slide-22
SLIDE 22

Thank You

John Fowler john.fowler@sun.com Executive Vice President, Systems Sun Microsystems, Inc.