Next Generation System Architecture Hp

Next Generation System Architecture Jerry Huck CTO – Business Critical Systems HP Chief Scientist © 2004 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice

The next big thing Price + Reliability + Security + Simplicity and manageability + Adaptability + Innovation + Connection

September 22, 2004

Draft

2

HP strategy

To offer a portfolio of products, services and solutions that are high-tech, low-cost and deliver the best customer experience

ch -te Hi gh

Customer

st

September 22, 2004

-co

– Carly Fiorina, HP CEO

w Lo

“Innovation at a price our customers can afford, delivered with an experience that sets us apart.”

Focused innovation that drives real value

Affordable technology that offers the best return on investment

Best customer experience

Draft

The best customer experience with HP and HP

3

The big shifts  All processes and content will be transformed from physical and static to digital, mobile and virtual.  The demand for simplicity, manageability and adaptability will change how customers work and organize, buy and use technology.  It’s a horizontal, heterogeneous, networked world. Standards are about connection and common language.

September 22, 2004

Draft

4

Mobility

Market drivers

Consumers •

Demand for new compelling services continues

Carriers •

Need new services (data + voice) to offset declining voice revenues

Business •

Tools to increase mobile productivity

•

Investments in wireless LAN technologies

22, 2004 • September Application

and device

Draft

5

Next Generation Technology Challenges and Opportunities - Itanium Systems - Blades - Virtualization - Storage Grids September 22, 2004

Draft

6

Itanium Systems – Building an enterprise system with standard processors September 22, 2004

Draft

7

Intel Itanium Architecture: Designed for Business Critical Next Enterprise Intel Computing ~6 Architecture Itanium ®

®

EPIC Instr / Cycle

Processor Age : 2+

~2

Performance

Superscalar

~.3

CISC

HP PA-RISC, Sun SPARC, IBM Power, MIPS, Alpha Age : 9-15+

Age : 20+

•

Built-in instruction-level parallelism

•

Issue ports and execution units support up to twice as many instrs/clock cycle

•

for 64-bits from the ground up • Architected for performance, scalability, and business critical availability

IBM 370, VAX 11

Time Performance through parallelism

Maturity curve narrows clock speed 22, gap over September 2004

Massive on-chip resources •

•

and more demanding workloads require new approach:

• Designed

~1

RISC

• Larger

Beating the memory latency gap & shorter Pipeline

128 general registers, 128 floating point registers, 8 branch ((vs 16 on x86) Fewer memory accesses (loads/stores) on complex workloads

Draft

•

Very large virtual and physical address spaces

•

Shorter memory pipeline

•

Latency avoidance

• •

Business Critical Availability •

Security: sophisticated ring protection and buffer overflow protection

Predication of instruction execution

•

Protected data paths

Data and control speculation

•

Failure mode analysis

Intel Itanium processor family and RoIT • Open

industry standard platform

− common platform for standard OS strategy − lowers HP’s R&D and support costs − ensure ISV adoption − ride industry standard ROI curve • Reduces

complexity

− provides legacy continuity − common management reduces IT TCO − common architecture provides flexibility for dynamic utility computing reconfigurations September 22, 2004

Draft

9

HP’s industry standards-based server strategy Moving to 3 leadership product lines – built on 2 industry standard architectures Future

Current HP NonStop server

Industry standard

Mips

HP NonStop server

HP Integrity server Itanium®

HP 9000 / e3000 server PA-RISC

HP AlphaServer systems Alpha

HP ProLiant server x86

September 22, 2004

Enabling larger investment in value-add innovation

Itanium®

Common technologies •

Adaptive Manageme nt

Itanium®

•

HP ProLiant server

Virtualizatio n

•

HA

•

Storage

•

Clustering

HP Integrity server

x86

10

Itanium® 2 & Xeon™ Processor MP Comparison Performance

Scalability

Reliability

Itaniu Xeon™ Characteristic MP m® 2 Error recovery on data bus ECC +30%-50% Internal soft +50%-100% 2005 error logic in ‘04 um i s n in ’07+ check Ita tform s Machine Check orm f Pla t a  l P Architecture ed s a -B w eon X ’s La l Bad data e r e t o In Mo  containment Cache 2005 Reliability ’07+ Lockstep ‘04   * For Enterprise & Technical Computing support Application Segments Memory SDEC, retry double  Today: bit 30-50% higher performance Memory spares     node Partitioning 2007:

Up to 2x performance at same platform price

600

+125%

500 400 300

+50%

200 100 0

4PTPCC

XeonMP Itanium2

node

High-end “RISC”-level RAS

Itanium’s EPIC Architecture: Highest Performance, Reliability, Scalability * Other names and brands may be claimed as the property of others. Enterprise Platforms

Group

32PTPCC

~2x Higher Scalability

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

11

Itanium-based servers will soon exceed IBM Powerbased servers in unit volume! (2004 or early 2005) IPF & RISC Server Volumes ( ku) 500 400 300 200 100 0 2002

2003

2004

2005 IPF

2006

2007

Power Source: IDC

September 22, 2004

Draft

12

“Inventing” a dual core Itanium® CPU

•

The standard Intel cartridge packaging is not at maximum density − CPU silicon chip is in a package, on a carrier board, with power on the end − The basic chip and package could be packed much more densely Industry-standard Itanium2® power pod (DC to DC power conversion)

Intel CPU “carrier” board

Intel CPU chip inside package HP CPU “carrier” board

HP external cache and controller chip

HP power solution (goes on top rather than on the end) September 22, 2004

Draft

13

sx1000 chipset error correction features PCI online add/replace

32 SDRAM DIMMs

Cell Board

DRAM chip-spare

2X MEM BUFFER

DIMM Address + Control Parity

Itanium® 2 Buses

2X MEM BUFFER

CPU

CPU

Socket*

Socket*

CPU Cache ECC

CPU

CPU bus ECC

Socket*

Socket*

ECC + single wire correct on M-links

2X MEM BUFFER

Cell Controller

Crossbar September 22, 2004

Draft

PCI-X PCI-X Host … Host Bridge Bridge 16 Links

…

Parity on PCI-X and “ropes”

I/O ECC PCI-X System + single Bus Adapter wire correct

ECC/Parity Internal to VLSI

2X MEM BUFFER crossbar hard partitioning

CPU

PCI/PCI-X Buses

I/O Subsystem

ECC + Single wire correct on crossbar links and VLSI

14

HP Integrity and HP 9000 Server Roadmap

Revision 4.5

Current offering

4-128P

2-16P 2-32P

HP Integrity Superdome

CPU: Itanium2 Madison 9M

HP Integrity rx7620-16, rx862032

CPU: Itanium2 Madison9M

1-8P

HP Integrity rx4640-8

1-4P

HP Integrity rx2600-2

1-2P

2004

DDRII

DDRII

2006

CPU: Montecito New Chipset CPU: Montecito New Chipset

PCI-E

PCI-E

CPU: Itanium2 Madison9M

CPU: Montecito New 8p Server & DDRChipset II PCI-E CPU: Montecito

CPU: Itanium2 Madison9M

CPU: Montecito New 4p Server & DDRChipset II PCI-E CPU: Montecito

CPU: DP Itanium

HP Integrity rx1600-2

Timeframes not to scale subject to Plans September 22, 2004

2005

July-04

All upgrades “in-box” except as noted New Chassis Intro.

Draft PCI-Express

CPU: Next Gen & Chipset PCI-EDDRII DP Itanium *Not available at initial processor release DDR-II Memory

15

Near Futures Summary •

Third cellular chipset planned for Montecito shipment in 2005 − Will also support PA-8900 on this new chipset

•

Increased Performance − − − − − −

•

1.3X CPU bus bandwidth increase 2X memory bandwidth increase 4X memory capacity 4X crossbar bandwidth increase 3X aggregate I/O bandwidth increase 2X I/O slot bandwidth increase (PCI-X 2.0 DDR)

Increased Reliability and Resiliency − − − −

Double DRAM sparing (tolerate failure of 2 DRAMs on a DIMM) Multi-path crossbar topology with self-healing links Redundant DC to DC converters and system clocks Greater fault protection within the CPU

September 22, 2004

Draft

16

HP server and storage portfolio

The world’s broadest, most robust enterprise offering

Database

Integrity Servers

NonStop Servers

NonSto p OS

Multi-OS

- Storage arrays and NAS - SAN infrastructure, software,media

Application

StorageWorks

- Tape and optical storage

ProLiant Servers HP9000 servers

Access

Alpha servers

Industry standard servers

Traditional RISC servers

Tru6 4

Management Services September 22, 2004

Draft

17

Blades – toward a utility computing fabric

September 22, 2004

Draft

18

HP Server Blade Portfolio Designed for adaptive, multi-tiered architectures new

1P Front End

2P Stateless New

ing m Co 005 2

2P Mid-Tier

4P Back-end

September 22, 2004

Draft

19

HP BladeSystem management:

New suite automates IT service delivery Optimizing HP Systems Insight Manager and iLO to create the hub of automated, virtualized HP BladeSystem management New plug-in controllers to SIM 1 Integrate 2 monitor/control

4

-Compute -Storage -Network -Power -Software

Virtualize

• Virtualization controller: Deploy, migrate, and manage heterogeneous virtual nodes • Automation controller: Workflow automation for policy-driven change of physical and logical resource pools

3

Adapt and Advise and change Automate Automated infrastructure control across the lifecycle

• Patch and scanning controller: OS and application vulnerability patch • Intelligent Networking: Failover network path for maximum availability and performance

• Dynamic power management: maximize energy efficiency & • Services to accelerate valuebalance and ensure success power draw • Affordable blade management software/hardware bundles

September 22, 2004

Draft

20

Drive Out Hard Costs ProLiant p-class blades

Acquisition Costs (based on 8 servers)

Traditional Rack-Mount Servers ProLiant DL360G3 w/ Ethernet and SAN

Cabling Connectivity (based on 40 servers)

40x3 network + 40x2 power = 200 cables

5x2 network + 4 power = 14 cables 93% reduction

Data Center space (density per 42U rack)

36 servers + Ethernet switches

48 - 96 servers + Ethernet switches 25 - 60% reduction

16.6kW*

12.1kW 27% savings Less PDUs (see whitepaper

Power and Cooling (based on 40 servers)

up to 19% savings

for whole list)

Failover servers

September 22, 2004

1+1redundancy

Draft

N+1redundancy ~50% savings

21

Virtualization – flexibility and agility in IT

September 22, 2004

Draft

22

servers will dramatically improve server utilization rates, increase server flexibility and reduce the overall spending required for servers…Virtualization should become an ongoing effort and part of the server strategy for every enterprise.” Gartner Server Virtualization Evolves Rapidly, 2003

Adaptive Enterprise vision Business and IT synchronized to capitalize on change Measure, assess and maintain a dynamic link between business and IT Architect and integrate heterogeneous IT environments

3.

Extend and link business processes across suppliers and customers

4.

EXTEND & LINK Suppliers

Manage and control business processes, applications and the whole IT environment

September 22, 2004

DraftUnder Non Disclosure until December 4, 2003

Employees

Customers

Applications ARCHITECT & INTEGRATE simplify, standardize, modularize, integrate

Virtualized resources

MANAGE & CONTROL assess, advise, act

2.

Business processes

MEASURE & ASSESS time, range, ease

1.

Infrastructure

page 24

Adaptive Enterprise design principles simplification

+

•

Reduce number of elements

•

Eliminate customization

Use standard technologies + standardization interfaces • •

Adopt common enterprise architecture

• •

Implement processes Break downstandard monolithic structures

•

Business processes

•

Create reusable components

•

Applications

•

Infrastructure

+ modularity

+ integration

Implement logical • architectures Manage the dynamic link between business + IT •

• September 22, 2004

Applied consistently across:

Connect apps + processes inside and outside DraftUnder Non Disclosure until December 4, 2003

page 25

Virtualization An approach to IT that pools and shares resources so utilization is optimized and supply automatically meets demand Business Supply

Demand

Information technology September 22, 2004

DraftUnder Non Disclosure until December 4, 2003

page 26

Just in time delivery changes the economics Just in case ERP

CRM

Web

Just in time ERP CRM Web

IT

Silos Server

Storage

•Pooled

Manufacturing

Network Software •Shared

September 22, 2004

Pooled Inventory Engine

Seats

Excess inventory Wheels

Chassis DraftUnder Non Disclosure until December 4, 2003

Engine

Seats

Wheels

Chassis page 27

Optimizing resources from desktop to datacenter VIRTUALIZATION INNOVATION TODAY

Business Value

Complete IT Utility Integrated Virtualization Element Virtualization Optimize utilization of server, storage or networking resources

Optimize all heterogeneous resources so supply meets demand in real time

Optimize environments to automatically meet service level agreements

Servers Storage NetworkSoftware

Strategic Importance September 22, 2004

DraftUnder Non Disclosure until December 4, 2003

page 28

Environment

Flexible Virtualization for HP-UX 11i Virtualization Techniques Process Management

Hard and Soft Partitioning

(single OS with or without psets)

(multiple OS images)

App A

On Demand

High Availability (across partitions or servers)

Disaster Tolerant

Partition A Partition Partition A

App B Partition B Partition Partition B

HP Process Resource Manager

HP nPartitions or HP Virtual Partitions

Instant Capacity On Demand (iCOD)

Metrocluster HP Serviceguard & Continentalcluster s

Intelligent Policy Engine: HP-UX Workload Manager (HP-UX WLM)

September 22, 2004

Single Server

Groups of Servers Draft

Geographically Dispersed Groups of

29

HP Integrity Virtual Machines … optimum utilization across Multi OS • Sub CPU virtual machines

app1

app2

app1

app2

app1

app2

app3

app4

app3

app4

app3

app4

app5

app6

OS (Linux)

OS (HP-UX 11i v2) OS (HP-UX 11i v2)

Intelligent Hypervisor

Memory I/O I/O

Hardware

I/O

2H 2005 September 22, 2004

Draft

with shared I/O • Runs on a server or within an nPar • Dynamic resource allocation built-in • Resource guarantees as low as 5% CPU granularity • OS fault and security isolation • Supports all (current and future) HP Integrity servers • Designed for multi OS – first on HP-UX 11i • VSE integration for high availability and utility pricing

30

Virtualization requires a new approach • Demand service levels not dedicated systems

People

• Transform “server huggers” into “service providers” • Measured and reward resource utilization and SLAs

Process

• Simplify and standardize IT processes based on best practices so you can replicate and automate • Monitor and charge back IT resources based on actual usage • Pool and share standard, modular IT resources

Technology

• Allocate resources dynamically to ensure SLAs are met • Optimize utilization and availability • Automate flow of resource supply to meet demand

September 22, 2004

DraftUnder Non Disclosure until December 4, 2003

page 31

Grids and Storage – another key part of a utility

September 22, 2004

Draft

32

Real-time scalability Scalable capacity

Evolving storage towards the Adaptive Enterprise

Storage Grid Network Storage DAS

Storage Grid enables  NEW information services Network Storage enabled Greater scalability Efficient resource sharing Higher system availability Shorter backup windows

Resource Real-timeefficiency information services September 22, 2004

33

HP StorageWorks Grid: A unique “array”

Hostconnected networks

High performance inter-smart cell networks

Storage clients September 22, 2004

34

HP StorageWorks Grid composition •

Smart cell hardware − − − −

•

Processor Cache Internal disks, tapes, etc. Off-the-shelf components

Smart cell software − − − −

Federates with peers Places data on internal devices Provides smart cell “personality” Responds to changes

Host Facing Network

Inter-smart cell Network

Host IOPs

Intercell IOPs

Processor, cache

Devices

 Service modules

 3rd party modules

• Redistributes data and workloads

− Hosting environment for 3rd party code •

 Plug-in environment

Smart cells are federated into domains  Common operating software September 22, 2004

35

Value •

HP StorageWorks Grids can exist on their own − No non-grid storage is needed − But customers will already have some

•

HP StorageWorks Grids can coexist with conventional storage − Customers can plug into existing storage networks

•

We expect customers to deploy multiple HP StorageWorks Grids − each HP StorageWorks Grid may contain multiple domains

•

Conventional arrays can be migrated into the grid

September 22, 2004

36

Conclusions

September 22, 2004

Draft

37

Future System Architecture

•

Customer

st

− No need to choose between highperformance computing and industrystandard economics − Unsurpassed agility and proven superior price/performance

co

Strategy that offers the best return on investment

w

•

Lo

− Select what’s best for you: open architecture approach across multiple operating environments.

te ch

HP shows proven technology innovation that drives real business results

Hi gh

•

Best Customer experience

Dedication to customer satisfaction that creates a superior customer experience − Collaborative customer relationship programs, enabling open and frequent communication − Clear roadmaps and commitments September 22, 2004

38

gh Hi

st co

Draft

w

September 22, 2004

Lo

−Jerry Huck CTO, HP Business Critical Servers Division. [email protected] , Tel:+1408-447-2429 −Harel Ifhar. HP Global Alliance Manager for Amdocs. [email protected] Tel:+972-524840-916 −Esteban Birenbaum. HP Technical Alliance Manager for Amdocs. [email protected] Tel:+972-52-4840918

te ch

Contact Info

Customer

Best Customer experience

39