Convergências, Integração e Desempenho no IBM System z Enterprise

Resumo

• João Paulo Cardani – Sou graduado em Engenheria Eletrônica pela UNICAMP, com especialização em Gestão Empresarial pelo Instituto de Economia da UNICAMP e licenciatura em Matemática pela UNAR.

• Entre 1987 e 1997 atue como Engenheiro de Produção, Manufatura e Desenvolvimento de Produtos de T.I. envolvido em projetos IBM no Brasil e EUA.

• Desde 1998 trabalho no departamento comercial da IBM em vendas de projetos de T.I. nas plataformas Mainframe e distribuída RISC, atendo os setores de Varejo, Governo e Datacenters.

• Sou Embaixador de Software Livre, ministrando palestras em diversas instituições de ensino.

• Meu foco desde 2008 é a comercialização de soluções e aplicações para Linux que rodam em servidores corporativos IBM Familia System z (descendente do Mainframe IBM, no mercado desde 1964).

© 2010 IBM Corporation

IBM zEnterpriseConvergência, Integração e Desempenho com IBM System z Enterprise

.João Paulo Cardani [email protected]ções Linux & IT Optimization

17 de Fevereiro de 2011

STG – Systems Technology Group

IBM zEnterprise | Enterprise Linux Server | Datacenter in a Box © 2010 IBM Corporation

TrademarksThe following are trademarks of the International Business Machines Corporation in the United States and/or other countries.

The following are trademarks or registered trademarks of other companies.* Registered trademarks of IBM Corporation

* All other products may be trademarks or registered trademarks of their respective companies.Notes: Performance is in Internal Throughput Rate (ITR) ratio based on measurements and projections using standard IBM benchmarks in a controlled environment. The actual throughput that any user will experience will vary depending upon considerations such as the amount of multiprogramming in the user's job stream, the I/O configuration, the storage configuration, and the workload processed. Therefore, no assurance can be given that an individual user will achieve throughput improvements equivalent to the performance ratios stated here. IBM hardware products are manufactured from new parts, or new and serviceable used parts. Regardless, our warranty terms apply.All customer examples cited or described in this presentation are presented as illustrations of the manner in which some customers have used IBM products and the results they may have achieved. Actual environmental costs and performance characteristics will vary depending on individual customer configurations and conditions.This publication was produced in the United States. IBM may not offer the products, services or features discussed in this document in other countries, and the information may be subject to change without notice. Consult your local IBM business contact for information on the product or services available in your area.All statements regarding IBM's future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only.Information about non-IBM products is obtained from the manufacturers of those products or their published announcements. IBM has not tested those products and cannot confirm the performance, compatibility, or any other claims related to non-IBM products. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products.Prices subject to change without notice. Contact your IBM representative or Business Partner for the most current pricing in your geography.

Cognos*DB2*Domino*GDPS*IBM*IBM logo*IMSParallel Sysplex*System z*System z9*

Adobe, the Adobe logo, PostScript, and the PostScript logo are either registered trademarks or trademarks of Adobe Systems Incorporated in the United States, and/or other countries.Cell Broadband Engine is a trademark of Sony Computer Entertainment, Inc. in the United States, other countries, or both and is used under license there from. Java and all Java-based trademarks are trademarks of Sun Microsystems, Inc. in the United States, other countries, or both. Microsoft, Windows, Windows NT, and the Windows logo are trademarks of Microsoft Corporation in the United States, other countries, or both.InfiniBand is a trademark and service mark of the InfiniBand Trade Association.Intel, Intel logo, Intel Inside, Intel Inside logo, Intel Centrino, Intel Centrino logo, Celeron, Intel Xeon, Intel SpeedStep, Itanium, and Pentium are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries.UNIX is a registered trademark of The Open Group in the United States and other countries. Linux is a registered trademark of Linus Torvalds in the United States, other countries, or both. ITIL is a registered trademark, and a registered community trademark of the Office of Government Commerce, and is registered in the U.S. Patent and Trademark Office.IT Infrastructure Library is a registered trademark of the Central Computer and Telecommunications Agency, which is now part of the Office of Government Commerce. © 2009 Gartner, Inc. and/or its Affiliates. All Rights Reserved

System z10Tivoli*WebSphere*z9*z10z10 BCz10 ECz196z/OS*z/VM*



AGENDA

2011 = Ano do Centenário da IBMhttp://www.ibm.com/ibm100/us/en/icons/index.html

IBM System zEnterprise™ - Revolução do Mainframe Convergência, Integração e Desempenho Mercado de Trabalho



IBM - CENTENÁRIA E INOVADORA

Uma Empresa que a 100 anos oferece Serviços e Produtos que contribuem com o sucesso de seus clientes

Faturamento Anual: $99 bilhões (2010)

Mercados Emergentes representa 21% e crescendo

Lucro Líquido: $ 18 bilhões

Acionistas: mais de 600 mil

Presente: em mais de 170 países do mundo.

Mais de 2.000 escritórios.

Mais de 400.000 funcionários

Planeta Mais Inteligente



A IBM em números hoje:

2ª Marca mais valiosa pela Consultoria Interbrand – interbrand.com 2009/2010

Lider Mundial em registro de Patentes por 18 anos consecutivos Sucesso em TI e Corporações devido a evolução tecnológica US$ 6 bilhões investidos anualmente em P&D 7 Laboratórios – 3000 cientistas – 5 prêmios Nobel 94 anos atuando no Brasil Mais de 17 mil Brasileiros



Algumas criações da IBM

RAMAC (1956) The First Magnetic Hard DiskThe world’s first hard disk drive was the size of two kitchen refrigerators side by side

IBM 1401: The Mainframe (1959)The first high-volume, stored-program, core-memory transistorized mainframe computer.

The Floppy Disk (1967)The IBM engineers who developed the floppy disk nevercould have dreamed that it would soon become instilled in the fabric of consumers' lives.It was originally designed for IBM's System/370 mainframe data loads

Magnetic Stripe Technology (1969)

Rise of the Internet1987 (NSFNET)



Voce já ouviu falar do IBM WATSON ?

Trata-se do mais novo computador da IBMque roda um software chamado Deep QA, desenvolvido

nos laboratórios da IBM, que a partir de sua experiencia e participação no jogo da TV americana chamado “Jeopardy”

pretende ajudar as pessoas e empresas a responderem questões envolvendo dados não estruturados mais

eficientemente



O que é um MAINFRAME ? Servidor Corporativo com grande capacidade para processar

TRANSAÇÕES.

Plataforma de Hardware e Software com maior índice de disponibilidade, segurança, escalabilidade e uso eficiente da energia.

Considerado o Servidor de GRANDE PORTE e de MISSÃO CRÍTICA.

Servidor usado por diversas empresas nos mais variados setores. Bancos, Telecom, Data Center, Públicos, Prestadores de Serviço, Games.

Atende a milhões de usuários com 1 único servidor e sistema operacional.

O primeiro ambiente a fazer VIRTUALIZAÇÃO, há mais de 40 anos.



Qual sua importância no mundo? Manter as operações e sistemas críticos das empresas funcionando 24

horas por dia, 7 dias por semana, sem interrupções não programadas.

Garantir que as transações, operações e fluxo de informações destas empresas sejam rápidas, eficazes e seguras.

Essa eficácia e disponibilidade contribui para a boa imagem da empresa na mercado, saúde do negócio, satisfação dos clientes e acionistas.

O sistema financeiro e de seguros mundial é suportado quase que na totalidade por Mainframes (95 % dos dados e operações críticas sendo processados na plataforma).*

No mundo 80% dos dados corporativos rodam no mainframe.*

* Fonte IDC e IBM



História do Mainframe Cenário 1 – Décadas de 60, 70 e 80 – Anos Dourados

– Lançamento, crescimento e consolidação da plataforma mainframe

– Ensino de Mainframe nas faculdades e abundância de especialistas mainframe, programadores COBOL.

– Momento de glória para o mainframe e facilidade para as empresas em encontrar recursos qualificados

Cenário 2 – Década de 90 – Decadência– Queda brusca das vendas de mainframe e muitas empresas migrando para PCs

– Popularização da micro-informática e obsolescência do mainframe

– Meio acadêmico e IBM deixam de lado a formação na área

– As empresas ficaram responsáveis por formar novos especialistas na plataforma

Cenário 3 – 2000 em diante – Renascimento– No final da década de 90 a IBM começa a investir remodelando o mainframe

– Em 2003 é lançado o Mainframe Charter – Compromisso público da IBM com investimento pesado da IBM nos próximos anos na Tecnologia, Valor e Comunidade da plataforma mainframe.

– Meio acadêmico volta a formar pessoas na tecnologia no mundo inteiro.

– Brasil é o 2º país que mais capacita pessoas em mainframe hoje.



Mainframe 1964

Cartão Perfurado



System IBM 3090 – refrigerado a água



Lançamentos 2010 – Mainframe System z196

Até 80 Processadores (para aplicação) 5.2 Ghz de Velocidade (Processador)Até 3.0 Terabytes de memória

3x mais memória 70% mais capacidade 1024 canais de I/O 6GBps por canalMenor consumo de Kw



Tamanho do Mainframe

1.77m



z196 – Dentro da Caixa

z196TLLB16

InternalBatteries(optional)

PowerSupplies

I/O cage

Fiber Quick Connect (FQC) Feature

(optional)

Processor Books, Memory, MBA and

HCA cards

2 x CoolingUnits

InfiniBand I/O Interconnects

2 x SupportElements

FICON & ESCON FQC

Ethernet cables for internal System LAN connecting Flexible Service Processor

(FSP) cage controller cards

I/O drawers



z196 Water cooled – Under the covers

z196TLLB17

InternalBatteries(optional)

PowerSupplies

I/O cage

Processor Books, Memory, MBA and

HCA cards

2 x Water CoolingUnits

InfiniBand I/O Interconnects

SupportElements

Ethernet cables for internal System LAN connecting Flexible Service Processor

(FSP) cage controller cards

I/O drawers



MAINFRAME – Book – Coração do Mainframe


IBM zEnterprise

Convergência, Integração e Desempenho



O que é virtualização ?O hardware x86 de hoje foi desenvolvido para executar um único sistema operacional e um único aplicativo, fazendo com que a maioria das máquinas seja subutilizada.

A virtualização permite executar várias máquinas virtuais em umaúnica máquina física, com cada VM compartilhando os recursos desse computador físico entre vários ambientes. Máquinas virtuais diferentes conseguem executar sistemas operacionaisdiferentes e vários aplicativos no mesmo computador físico.

Enquanto outras empresas ainda estão dando os primeiros passos em direçãoà virtualização, a VMware já é a líder do mercado. Nossa tecnologia é comprovada na produção e utilizada por mais de 170 mil clientes, incluindo 100% das empresas da Fortune 100.

Fonte: Páginas internet de vários fornecedores



Cinco principais razões para adotar virtualização

1) Obtenha mais de seus recursos existentes consolidação de servidores.

2) Reduza os custos de data center diminuindo sua infraestrutura física: exigências de energia e refrigeração.

3) Aumente a disponibilidade do hardware e de aplicaticos continuidade dos negócios

4) Obtenha flexibilidade operacional resposta rápida a novas demandas

5)Aprimore a capacidade de gerenciamento e a segurançagerenciamento centralizado

Fonte: Páginas internet de vários fornecedores



Porque a virtualização e a consolidação são importantes ?

Instalações– Espaço

– Energia

– Resfriamento

Hardware– Servidores

– Armazenamento

– Rede

– Switches e Roteadores

Software– Licenças– Suporte– Manutenção

Administração– Site– Servidores– Software– Aplicações– Dados



Conceitos de virtualização

Recursos Virtuais- Para recursos reais: mesmas interfaces/funções, atributos diferentes- Pode ser parte de um recurso físico ou múltiplos recursos físicos

Virtualização- Cria recursos virtuais e em seguida, “mapas” para os recursos reais.- Principalmente realizados com software ou firmware.

Recursos- Componentes arquitetados com interfaces / funções- Pode ser centralizada ou distribuída. Normalmente física.- Exemplos: memória, discos, redes, servidores.

Separa a apresentação de recursos para os usuários de recursos reais.Agrega “pools” de recursos virtuais para alocação para os usuários .



Evolução da Tecnologia de Virtualização

VP/CSS

VM/370

VM/SP

VM/XA

VM/ESA

z/VM

1967 1968 1972 1980 1988 1990 2000

Desenvolvida para explorar a tecnologia de hardware Constantamente melhorada para atender os requisitos de negócio Inovação contínua para adaptação Maximiza a produtividade

CP-40/CMS

CP-67/CMS



O que é consolidação ?

Em TI, consolidação, é a capacidade de se compartilhar umainfraestrutura (fonte de alimentação, cabeamento de rede, roteador,etc) e recursos computacionais (processadores, memória, disco, etc)com a instalação de servidores em um hardware único comgerenciamento centralizado, seguro e controlado extraindo o melhor detodos os componentes envolvidos.

HW 1

Server1

Server2

Server3

Server4

HW 2

Server21

Server22

Server23

Server24

HW 3

Server31

Server32

Server33

Server34

HW 4

Server41

Server42

Server43

Server44

...

.........



Tecnologia da Informação hoje : Limitações

Complexidade e Limitações do Ambiente Atual Muitas camadas/nós de recuros independentes resources conectados via a rede

corporativa System management information typically not end-to-end view Automation Policies are limited to tier/node boundaries Redundancy is pervasive for Operational staff, HW, Software and policies across

architectures Managing this complexity now consumes the majority if IT budgets

Information Technology today is limited by the technology and architecture configurations available

DS Servers

LAN Servers

SSL/XMLAppliances

CachingAppliances

RoutersSwitches

FirewallServers

Security/Directory Servers

Application Servers

File/Print ServersBusiness Intelligence

Servers

Web Servers

System z



Em seis anos, o consumo de energia de um servidor saiu de 8 watts para mais de 100 watts para cada U$ 1.000 de tecnologia

Em média, para cada 100 unidades de energia gastos num datacenter, apenas 3 unidades são usadas para processamento. E mais da metade da energia é usada para o resfriamento do ambiente.

Em ambientes de computação distribuída, 85% da capacidade dos servidores está ociosa.

Quanto de cada servidor x86 está sendo efetivamente utilizado ?

Quanta energia está sendo gasta para resfriar estes servidores ?

Quantas pessoas são necessárias para gerenciar esses servidores físicos ?

Por quanto tempo você continuará a executar desempenhar o trabalho lento e massivo de cuidar de grande quantidade de servidores físicos ?

É tempo de começar a pensar “diferente” a infra-estrutura



Os desafios dos DataCenters Globais

DataCenter VERDE Computação em Nuvem Pública/Privada Software as a Services (SaaS) Processamento de Aplicações para Terceiros Certificação Digital / Entidade Certificadora Segurança de Informação e Dados Automação e Provisionamento de Recursos de TI Consolidação e Virtualização de Servidores Ambientes de Contingencia e Recuperação de Desastres Provas de Conceito, Homologação de Aplicações, SLA Gerenciamento Integrado de Sistemas Heterogeneos Gestão de TCO e TCA



O que o Mainframe IBM pode oferecer

Mais de 47 anos de experiência e inovação tecnológica em hardware e software RAS – Confiabilidade, Disponibilidade e

Manutenção Mais de 40 anos de experiencia em virtualização,

consolidação e gerenciamento dinamico de recursos SLA consistente Ciclo de vida e compatibilidade entre gerações de

produtos consistente TCO



Mainframe era um equipamento que executava tarefas complexas da forma usual no mundo real:

Dividindo trabalho entre vários grupos de especialistas

System z Enterprise é um mainframe que leva essa abordagem para o mundo distribuído, combinando:

• um hardware poderosíssimo: z196 + zBX • um hypervisor levíssimo: o z/VM• um gerenciador de cargas heterogêneas: o zManager

O que o era um Mainframe em 1964 ?

O que é o System z Enterprise ?



Next Generation Data Center (NGDC)

O sonho de consumo dos clientes Aderencia as leis e regulações de mercado Os caminhos percorridos até 2010 A jornada a seguir para os próximos 5 anos



INOVAÇÃO

SUSTENTABILIDADESUSTENTABILIDADE

SIST. INTELIGENTESSIST. INTELIGENTES

SEGURANÇASEGURANÇA

Design de um Smarter “Next Gen” Data Center



As Forças de Mercado são Reais

Analise de ações baseada na sustentabilidade anual

Classificação Ambiental para mais

de 2.000 empresas públicas

Analia a performance ambiental de 1.200 empresas

Representa 385 investidores globais com mais de US$57 trilhões em assets. Reporta emissões de CO2 de 1.550 grandes corporações ao redor do globo

Regulatory Compliance *The European Union (EU) introduced a voluntaryCode of Conduct for energy efficiency in datacenters in 2009. The uptake has been slow, partiallybecause of the bureaucracy of the EU, but alsobecause other countries such as Australia and the U.S.are looking to introduce such codes. One benefit is thatan integrated energy management architecture, whilenot necessity to obtain certification, is encouraged andwill make renewing such a commitment easier. Asecondary benefit is around carbon disclosure. Therequirement to report the carbon emissions fromlarge data centers will grow during the next fiveyears. Gartner believes that in countries such asthe U.K., the government will ask large data centersto voluntarily disclose their carbon emissions by2013. In the U.S., the Energy Star Program is beingextended to servers, and will include storage andnetworking devices during the next 24 months.

* Gartner ResearchPublication Date: 26 April 2010/ID Number: G00175179



Hardware

Consolidação/VirtualizaçãoSystems Management

Data Centers Verdes

Data Centers Verdes Avançados

Infraestrutura verde de ponta a ponta



Política de segurança da informação Organização da segurança da informação Gestão de ativos Segurança em Recursos Humanos Segurança física e do ambiente Gerenciamento das operações e

comunicações Controle de acessos Aquisição, desenvolvimento e

manutenção de sistemas de informação Gestão de incidentes de segurança da

informação Gestão da continuidade do negócio Compliance

Fonte: ABNT NBR 17799/2007 - 27002

Segurança Integrada



Tivoli Green Management(Monitor, Measure and Manage)

IBM Systems Director Active Energy Manager

Tivoli Green Management(Monitor, Measure and Manage)

IBM Systems Director Active Energy Manager

Ativos de TI

Servidores e Storage de 3os

Ativos de Infra de Facilities

Sistemas Prediais(HVAC, alimentação,

iluminação, segurança)

Ativos de Infra de Datacenter

Datacenter Infra(UPS, PDUs)

Visualização Integrada

Sistemas Inteligentes: Gestão de Energia



Impacto no Negócio Custo da disponibilidade

Time To Recovery

$

8hrs

24hrs

48hrs

Days Weeks72hrs

Financial Exposure

Long Term Financial Impact /Customer Base

Lost

Determine The Impact

Productivity / Quality Losses

Short Term Financial Impact / Customer Base

Affected

Time To Recovery

Hot SiteData Center

$

AvailabilityInvestment

Cold SiteData Center

Mirrored or Replicated

Data

8hrs

24hrs

48hrs

Days Weeks72hrs

Determine The Cost Of

Availability

versus

*Fonte: Jim Szel, Dennis Mullin, – both of Syska Hennessy Group – 3/30/05*Fonte: Jim Szel, Dennis Mullin, – both of Syska Hennessy Group – 3/30/05

Resiliência do Negócio x Resiliência do Data CenterResiliência do Negócio x Resiliência do Data Center



*Fonte: Jim Szel, Dennis Mullin, – both of Syska Hennessy Group – 3/30/05*Fonte: Jim Szel, Dennis Mullin, – both of Syska Hennessy Group – 3/30/05

Erro Humano Falta de Metodologia Procedimientos Incorretos Desenho Pobre Redundância Inadequada Falta de Manutenção

Falha nos Equipamentos Ativos Falha nos Provedores Terrorismo Cibernético

Imprevisíveis(33%)*

Previsíveis(67%)*

Resiliência do Negócio x Resiliência do Data CenterResiliência do Negócio x Resiliência do Data Center



Aumento na Demanda de Potência e Refrigeração

Mudança na Dinâmica dos Custos

Infraestrutura do Data Center x Novas Necessidades

O Desafio dos Data CentersO Desafio dos Data Centers



A Evolução do Data Center A Evolução do Data Center

Rel

evan

cia

e C

ontr

ole

Rel

evan

cia

e C

ontr

ole

Evolução da ArquiteturaEvolução da Arquitetura

Data Center 1.0Data Center 1.0Centralizado

Fechado/Vertical/

CENTRALIZADO

Data Center 2.0Data Center 2.0Client-Server e Distributed

Computing

DESCENTRALIZADO VIRTUALIZADO

Data Center 3.0Data Center 3.0Service Oriented e Web

2.0 BasedHibrido/Integrado/

Consolidar

Virtualizar

Automatizar



Arquitetura de Aplicação 3-Camadas : exemplo típico no DataCenter

Web Servers

Application Servers

Database ServersRouter

Switch

Firewall

DataCenter



Transaction Processing & Database

Business Applications Web, Collaboration and Infrastructure

Analytics

Application Database Data Warehousing Online Transaction

Processing Batch

Data Mining Applications Numerical Enterprise Search

Enterprise Resource Planning Customer Relationship

Management Application Development

Systems Management Web Serving/Hosting Networking File and Print

zEnterprise: O Sistema dos SistemasGerenciando aplicações multi-camada pelo Enterprise


IBM zEnterprise | Enterprise Linux Server | Datacenter in a Box © 2010 IBM Corporationz196TLLB43

IBM System z family

Announce 2/08 – Server w/ up to 77 PU cores5 models – Up to 64-wayGranular Offerings for up to 12 CPsPU (Engine) Characterization

– CP, SAP, IFL, ICF, zAAP, zIIPOn Demand Capabilities

– CoD, CIU, CBU, On/Off CoD, CPEMemory – up to 1.5 TB for Server and

up to 1 TB per LPAR– 16 GB Fixed HSA

Channels– Four LCSSs– 2 Subchannel Sets– MIDAW facility– 63.75 subchannels– Up to 1024 ESCON® channels– Up to 336 FICON® channels– FICON Express2, 4 and 8– zHPF– OSA 10 GbE, GbE, 1000BASE-T– InfiniBand® Coupling Links

Configurable Crypto Express3Parallel Sysplex clusteringHiperSockets™ – up to 16Up to 60 logical partitionsEnhanced AvailabilityOperating Systems

– z/OS, z/VM, z/VSE™, TPF, z/TPF, Linux on System z

IBM System z10 EC (2097) IBM System z10 BC (2098)

Announced 10/08 – Server w/ 12 cores Single model – Up to 5-way CPsHigh levels of Granularity available

– 130 Capacity IndicatorsPU (Engine) Characterization


– CoD, CIU, CBU, On/Off CoD. CPEMemory – up to 256 GB for Server

– 8 GB Fixed HSAChannels

– Two LCSSs– 2 Subchannel Sets– MIDAW facility– 63.75 subchannels– Up to 480 ESCON channels– Up to 128 FICON channels– FICON Express2, 4 and 8 – zHPF– OSA 10 GbE, GbE, 1000BASE-T– InfiniBand Coupling Links

Configurable Crypto Express3Parallel Sysplex clusteringHiperSockets – up to 16Up to 30 logical partitionsEnhanced AvailabilityOperating Systems

– z/OS, z/OS.e, z/VM, z/VSE, TPF, z/TPF, Linux on System z

Announce 7/10 – Server w/ up to 96 PU cores5 models – Up to 80-wayGranular Offerings for up to 15 CPsPU (Engine) Characterization


– CoD, CIU, CBU, On/Off CoD, CPEMemory – up to 3 TB for Server and

up to 1 TB per LPAR– 16 GB Fixed HSA

Channels– Four LCSSs– 3 Subchannel Sets– MIDAW facility– 63.75 subchannels– Up to 240 ESCON channels– Up to 288 FICON channels– FICON Express4 and 8– zHPF– OSA 10 GbE, GbE, 1000BASE-T– InfiniBand Coupling Links

Configurable Crypto Express3Parallel Sysplex clusteringHiperSockets – up to 32Up to 60 logical partitionsEnhanced AvailabilityUnified Resource ManagerOperating Systems

– z/OS, z/VM, z/VSE, z/TPF, Linux on System z

Announce 7/102 models zBX Racks with:

– BladeCenter Chassis– N + 1 components– Blades– Top of Rack Switches– 8 Gb FC Switches– Power Units– Advance Management Modules

Model 001 for z107 to 56 Blade

– IBM Smart Analytics Optimizer Model 002 for z1960 to 112 Blades

– IBM Smart Analytics Optimizer Solution

– POWER7 Blades– System x Blades*– WebSphere DataPower

Appliance*

*All statements regarding IBM future direction and intent are subject to change or withdrawal without notice, and represents goals and objectives only.

IBM zEnterprise 196 (2817) IBM zEnterprise Blade Extension (2458)



Re-write the rulebook and set new standards for business-centric IT with IBM System z, to be the world’s premier workload-optimized platform for enterprise applications.

Nossa Visão:Deliver the best of all worlds - Mainframe, UNIX®, x86 and single function processors - integrated in a single system for ultimate flexibility and simplicity to optimize service, risk, and cost across multiple heterogeneous workloads.

O que é o IBM zEnterprise System?



zEnterprise 196

zEnterpriseUnified Resource Manager

– Unifies resources, extending System z qualities of service across the infrastructure

– Install, Monitor, Manage, Optimize, Diagnose & Service

Optimizers– Workload specific

accelerators to deliver a lower cost per transaction, appliance for example IBM Smart Analytics Optimizer

IBM Blades– Runs app unchanged and

supports what you know. Logical device integration between System z and distributed resources

– The industry's fastest and most scalable enterprise server

– Ideally suited for large scale data and transaction serving and mission critical enterprise applications

zEnterprise BladeCenter Extension

HMHMCC

Integration and centralized management by System z

IBM zEnterprise SystemA integração com Liderança Tecnológica


IBM zEnterprise | Enterprise Linux Server | Datacenter in a Box © 2010 IBM Corporation© 2010 IBM Corporation

5.2 GHz superscalar processorUp to 96 Cores, 1 to 80 configurable for client use

Up to 3 TB RAIM memory

Over 100 new instructions 1.5MB L2 Cache per core, 24MB L3 Cache per processor chip Cryptographic enhancements

Optional water cooling

60%with

More capacity

60%Up to

Improvement in performance

Improvement in performance

35%for

Less cost

Up to40%

Dramatic improvement over IBM System z10™:

For Linux For z/OS

Sem aumento de consumo de energia

Mais desempenho computacional com novos softwares

IBM zEnterprise 196: O coração da nova máquina O mais rápido e escalavel servidor de grande porte da industria



New 5.2 GHz Quad Core Processor Chip boosts hardware price/performance

– 100 new instructions – improvements for CPU intensive, Java™, and C++ applications

– Over twice as much on-chip cache as System z10 to help optimize data serving environment

– Out-of-order execution sequence gives significant performance boost for compute intensive applications

– Significant improvement for floating point workloads Performance improvement for systems

with large number of cores – improves MP ratio Data compression and cryptographic

processors right on the chip

z196 - IBM Leadership Technology at the Core



Optimizers• IBM Smart Analytics

Optimizer• DataPower®

Select IBM Blades• BladeCenter PS701 Express• System x1

Até 4 racks compartilhados , capacidade para 112 blades Configuravel com alta disponibilidade Rede Privada Segura entre zBX e z196 para

dados e suporte– Fast 10 Gb Ethernet connection to the data– Less latency – fewer ‘hops’ to get to the data an no need for

encryption / firewall– Traffic on user networks not affected.

Nivel de suporte padrão System z – Problem reporting, hardware and firmware updates

IBM zEnterprise BladeCenter Extension (zBX)

… managed by the zEnterprise Unified Resource Manager

1 All statements regarding IBM future direction and intent are subject to change or withdrawal without notice, and represents goals and objectives only.

Extendendo a Magia do System z para as Plataformas Heterogeneas ……



Service Managementfrom Tivoli®

Provide the visibility, control and automation needed to design, deliver, and manage differentiated services

Enable innovation and maximize performance with real-time service visibility and intelligence

Enforce security policy compliance

Centrally manage and protect access to services and resources

Unified Resource Manager

Hypervisor management and creation of virtual networks

Operational controls, service and support for hardware / firmware

Network management of private and secure data and support networks

Energy monitoring and management

Workload awareness and platform performance management

Virtualization management – single view of virtualization across the platform

Integração de Serviços e Gestão da PlataformazManager : Unified Resource Manager



Visão geral da zEnterprise

Blade Virtualization

Linux on System x 1

zBX

Blade Virtualization

AIX on Power

Application Server Blades

Blade HW Resources

Optimizers

Futu

re O

fferin

g

Smar

t Ana

lytic

s O

ptim

izer

Futu

re O

fferin

g

Customer Network Customer Network

z HW Resources

System z PR/SM™

Sys

tem

z H

ardw

are

Man

agem

ent C

onso

le

System z Host

z/OSzTPF

z/VSE™Linux

z/VM

Support Element

Linux

with

Uni

fied

Res

ourc

e M

anag

er

Private High Speed Data NetworkPrivate Management Network Ensemble Management

Firmware

Futu

re O

fferin

g

Private Data Network


3 anos, 5.000 profissionais alocados, 31 milhões de horas, 1,5 bilhão de dólares



Arquitetura de Aplicação 3-Camadas : Alternativa com System zEnterprise

Firewall

IEDN

Web Servers

Application Servers

Database Servers

zBX z196

IEDN = IntraEnsemble Data Network

Next Gen DataCenter

Storage

IBM SWG Lab Services and Training


zBX (nominal)

52



zBX (actual)

53



zBX (actual)• Frames B, C, D, and E• Top-of-rack (TOR) switch

– 2 Management• 40 port• 1 Gbps

– 2 Data• 16 port• 10 Gbps

• Optional Storage• Blades

54



HMC Connection to CEC (nominal)

zEnterprise BladeCenter Extension

Installation Manual

2458-002

GC27-2610-0055



HMC Connection to CEC (actual)

56



HMCHMC



Energy

ManagementHardware Management and Platform Management

HMCHMC

Hypervisors Energy

Networks

Performance

VirtualServers

Operations

Hypervisor Management Integrated deployment and configuration

of hypervisors Hypervisors (except z/VM) shipped and

serviced as firmware Management of ISO images Creation of virtual networks Manage and control communication

between virtual server operating systems and the hypervisor

Operational Controls Auto-discovery and configuration

support for new resources Cross platform hardware problem

detection, reporting and call home Physical hardware configuration,

backup and restore Delivery of system activity using new

user interface

Network Management Private, secure and physically isolated data and

service networks

Energy Management Monitoring and trend reporting of CPU

energy efficiency Static power saving Ability to query maximum potential

power

Single view of virtualization across platforms. Ability to deploy multiple, cross-platform virtual

servers within minutes Management of virtual networks including access

control

Virtual Server Lifecycle Management

Wizard-driven management of resources in accordance with specified business service level objectives

HMC provides a single consolidated and consistent view of resources

Monitor resource use within the context of a business workload

Define workloads and associated performance policies.

Workload Awareness and Platform Performance Management



A zEnterprise ensemble is a collection of 1 to 8 z196 CPCs with/without zBX managed collectively by the Unified Resource Manager as a single logical virtualized system using the HMC A zEnterprise node is a z196 CPC with 0 to 4

racks up to 2 BladeCenters per rack – zEnterprise nodes are deployed within a single site

– A zEnterprise node can be a member of at most one ensemble

z196 CPCs are deployed within a single site Blade based fit-for-purpose Solutions Integrated Advanced Virtualization Management Implements well-defined external interface to

Data Center Service Management functions Virtual Resource Management and Automation z10 can access the Optimizers, but can’t be part

of the managed ensemble

What is a zEnterprise Ensemble?

HM

C –

Uni

fied

Res

ourc

e M

anag

erzEnterprise NodezEnterprise Ensemble



Data-POWER7

- MCPDWA MCPDWAMCPMCPData-

Power

kvm

DWA

kvm

zEnterprise internal network


LinuxLinuxLinux Linux

LinuxLinuxLinux

Linux

MCPMCPData

kvm

ISAOPT

kvm

Firmware


z196 Ensemble Network

LinuxLinuxLinux AIX

LinuxLinuxLinux

Linux

kvmLinuxLinuxLinux

Linux

Bla

de V

irtua

lizat

ion

LinuxLinuxLinux

Linux

kvm

LinuxLinuxLinuxLinux B

lade

Virt

ualiz

atio

n

AIXAIXAIXAIX

Dynamic Virtual Network Provisioning Supporting Image Deployment and Mobility Local Network Optimizations Improved Security and Isolation Trusted Compute Base in Firmware Managed within the Ensemble

z/VM

LPARz/VM

LPAR

Linu

x

VIOP

z/O

S

z/VM

LPAR

Linu

xLi

nux

Linu

x

Linu

xLi

nux

z/V

SE

z/O

S

z/O

S

z/O

S

IBM x86 Blades*

POWER7 Blades

DataPower XI50zz196 System(s)IBM Smart Analytics

Optimizer

zEnterprise Ensemble – Logical view




Support LAN (INMN):– Hardware and Firmware

Service– Policy management interface– zBX managed by a single CPC

Data LAN (IEDN):– Access path from host OS

images– Independent of Service LAN– Multiple CPCs can attach

Example of 4 node z196 Ensemble

Shared zBX in an Ensemble



Define Workload and Performance Policy

z196PR/SM

z/OS

DB2

z/VM

Linux

WAS

zBX

Smart Analytics Optimizer

Power blade

AIX

Apache

INMN

PowerVM

IEDN

Performance Policy

HMC

PSCN



External Customer Data Network

Connecting the Customer external Data Network to the intraensemble data network – using unique VLANs

Virtual Server72B

Hypervisor

TCPIP1 (z/OS1)

OSX

Virtual Server Virtual Servers

Top of Rack

OSD

External Customer

Data Network

IP Filtering

IP Filtering

MAC Filtering

1. Enter through z196 via z/OS via an OSD.

2. Enter through a External Router to the zBX TOR.

1

2Router

Router

VLAN Enforcement

VLAN Enforcement

* and* Network Access Control through RACF®

Linux 51

Linux 55

VLAN Enforcement

VLAN Enforcement

VLAN AVLAN A

VLAN B

VLAN B

IEDNzBX



Multiple Enterprise Nodes – Sharing zBX(s) within single Node

System z

OSA (OSX)

BCBCBCBCTOR

zBX Racks

ESM ESM ESM ESM

System z

OSA (OSX)

BCBCBCBCTOR

zBX Racks

ESM ESM ESM ESM

zEnterprise

Node 1

(with zBX)

zEnterprise

Node 2

(without zBX)

Configuration - Multiple Enterprise Nodes (sharing zBX rack(s))

System z

VSwitch

LP 1

z/OS 1

LP 2

z/OS 2

LP 3

z/OS 3

LP 4

z/OS 4

LP 5 (zVM 1)

VS1 VS2 VS3 VS4

OSA (OSX)

intraensemble

data network

SE 1

SE 2



Banking Insurance Retail Healthcare Public Sector

Core Banking Internet Rate Quotes

On-line Catalog

Patient Care Systems

Electronic IRS

Wholesale Banking –Payments

Policy Sales & Management

(e.g. Life, Annuity, Auto)

Supply Chain Management

On– line Claims

Submission & Payments

Web based Social

Security

Customer Care & Insight

Claims Processing

Customer Analysis

zEnterpise é Ideal para acomodar aplicações das mais variadas Industrias


IBM zEnterprise | Enterprise Linux Server | Datacenter in a Box © 2010 IBM Corporationz196TLLB661 All statements regarding IBM future direction and intent are subject to change or withdrawal without notice, and represents goals and objectives only.

DistributedSystems

Select IBM Blades in zBX

z/OS

TCA Focus

TCO Focus

Linux on z/VM®

Expanded ISV support for enterprise applications

Targeted for applications that interact with mainframe data and transactions

Provisioned and managed by System z

CPU-bound applications interacting with System Z z/OS or Linux

Extreme consolidation of servers and networking

Superior levels of virtual server provisioning, monitoring and workload management

Industry-best virtual I/O bandwidth and reliability

Fewer components and reduced complexity

System z qualities of dynamic resource management and capacity-on-demand

Seamless integration with z/OS backup and disaster recovery solutions

Silo managed islands of computing

Less dynamic than z virtualization

Minimal resource sharing with z resources

Extreme scalability and performance for transaction processing and data serving

High availability and cross-system scalability with Parallel Sysplex® and GDPS®

Leading policy-based capacity provisioning and workload management

Pervasive, high-performance security support

LOWER HIGHERSCALABILITY, SECURITY, DYNAMIC WORKLOAD MANAGEMENT

SCALABILITY, SECURITY, DYNAMIC WORKLOAD MANAGEMENT

zEnterprise System

Service levels to match your business needs Increased flexibility for your multi-architecture strategy when data is on z/OS



Solução SAP

AIXSAP App Servers

DB2 Connect

Linux* on x

SAP App ServersDB2 Connect

SAP App Servers

Windows

DB2 Connect

DB2 Data Sharing

Central Instance Services

z/OS

SYSPLEX

LPAR, z/VM

SAP App ServersDB2 Connect

Linux on z

Hipersockets

Networking / Messaging

SAP Database Servers

zEnterprise will:- Improve management

and efficiency of non-z systems supporting a zSAP DB

- Increase uptake of DB Server on z/OS –dragging incremental volumes for all supported application platforms

- Improve competitive posture for customers who do not run application on an IBM platform today

- Improve competitive posture against HP for SAP Standalone Engines.

z196 e zBX

zNext will cover most of the application server computing requirements for today‘s zSAP customers - plus opportunity to expand.

Private Network

Adobe Document

Services x,p

TREX x(search/classification)

liveCache x,p

MDM x,p

Standalone Engines

Business Warehouse Accelerator

SAP Appliances„Black Box“

X86 Blade Center

Enterprise Service

Repository

*All statements regarding IBM's plans, directions, and intent are subject to change or withdrawal without notice. Any reliance on these Statements of General Direction is at the relying party's sole risk and will not create liability or obligation for IBM.



Cloud Computing on zEnterpriseBasic Architecture design Management LPAR provides a

“managed from” infrastructure, consisting of Linux (SUSE) guests running TSAM and OMEGAMON

– Rapid automation and services lifecycle management for z/VM based Linux cloud services

Workload LPAR provides the “managed to” environment, supporting the customer defined cloud images

– Supports Linux (SUSE & Redhat) & z/OS workloads support under z/VM

– A sample workload is provided

z/VM

Tivoli Services

Automation ManagerM

anagement LP

AR

Workload LP

AR

ImageA

ImageB

Web 2.0 Service Portal

Workload images

z/VM

OMEGAMON

User request

Request serviceProvision image



Segurança

Arquiteturade I/O

Capacidadesob demanda

Provisionamentode Servidores

Licenciamentode Software

Gerenciamento deSistemas

Arquiteturade processador

Gerenciamentode Recursos

Particionamento eVirtualização

Compartilhamento de Servidores e VirtualizaçãoAfeta várias áreas de TI . . .Em vários aspectos da arquitetura !

RASReliability, Availability & ServiceabilityNão basta ser estável, precisa ter flexibilidade para mudar sem haver interrupçõesComo modificar um servidor entre 300, sem comprometer os outros ?

• No mundo físico é fácil . . . E no mundo virtual ?



Exemplo de Virtualização – picos coincidentes e espaço virtual

00:10:02 08:10:01 16:10:01 00:10:01 08:10:01 16:10:01 00:10:01 08:10:01

Time Of Day

0

200

400

600

800

1000

1200

MIP

S

Server1Server2Server3Server4Server5Server6Server7Server8Server9Server10Server11Server12Server13Server14Server15Server16Server17Server18Server19

MIPS Consumed by Serverfor 24 hrs each day in 30 min int

Servidores OracleDados coletados durante 3 dias

De 19 servidores Linux x86 (49 Cores)



Core's Clock Watts BTU/hHP SuperDome 128 1,6 19.600 83.300Sun M9000 128 2,4 31.600 145.357IBM Nehalem 512 2,9 49.920 96.896IBM p595 64 5,0 28.300 83.600IBM z10 77 4,4 23.300 81.200

Estimativa para virtualização: 1 para 4Além de centenas de core's especializados

30% Overhead

Virtualizador

30% segurança

40% Úteis

30% Overhead

Virtualizador

30% segurança

40% Úteis

30% Overhead

Virtualizador

30% segurança

40% Úteis

30% Overhead

Virtualizador

30% segurança

40% Úteis

30% Overhead

Virtualizador

30% segurança

40% Úteis

30% Overhead

Virtualizador

30% segurança

40% Úteis

30% Overhead

Virtualizador

30% segurança

40% Úteis

30% Overhead

Virtualizador

30% segurança

40% Úteis

Virtualização em x86:Perda de até 60% da capacidade de cada server

20% Overhead Virtualizador

50% Úteis

30% segurança


50% Úteis

30% segurança


50% Úteis

30% segurança


50% Úteis

30% segurança

Virtualização em RISC:Perda de até 50% da capacidade de cada server

Virtualização em Mainframe: mais de 90% da capacidade é realmente utilizável pelos servidores virtuais Linux

Vantagens da Virtualização em System z: economia de energiaVIRTUALIZATION

Menos Energia Menos Calor

Mais Capacidade

MAIS VERDE !!



Virtualização em System zConclusão:

Maior capacidade aproveitada, menor overhead, casamento entre HW e SW

Seu GHz é mais útil !

(e por GHz, entenda-se kW/h, BTU, etc)



IBM System z: System Design Comparison

z196TLLB73

Balanced SystemCPU, nWay, Memory,

I/O Bandwidth*

Memory

System I/O Bandwidth

Processors

PCI for1-way

1.5 TB**

64-way

920

288 GB/Sec*

80-way

3 TB** 1202

172.8 GB/sec*

600512 GB

54-way

96 GB/sec

450256 GB

32-way

24 GB/sec

30064 GB

16-way

z10 EC

z9 EC

zSeries 990

zSeries 900

z196

* Servers exploit a subset of its designed I/O capability** Up to 1 TB per LPARPCI - Processor Capacity Index



Conceitos



Microprocessador6502 (6800), z80, 8088, 8086Basic, CPM, MSDOS, Xenix

Arquitetura de Sistemas – o Início

Computador de baixo custo

Todas as funções de um computador

emuladas por software

I/O DeviceDrivers Criptografia, etc

Código de Aplicação

OS e Gerenciamentode Recurso (não ‘s)

Monotarefa eMonousuário



Arquitetura de Sistemas – Multitarefa

ProcessosAplicação 1Aplicação 2Aplicação 3Aplicação 4

Aplicação N

app1

app2

app3

app4

appN. . .

app1

Tempo

• Time Sharing• Time Slice• Modo Usuário• Modo Supervisor (Sys)• Troca de Contexto

Medido em laboratório: um servidor RISC a 100% com aplicação comercial: 43% do tempo em modo supervisor

User Mode:

Dis

k w

Dis

k r

LAN

w

Dis

k r

LAN

w. . . Dis

k w

Sys Mode:



Arquitetura de Computadores – Comparativo de Processadores

• Cache L3 “on-chip”:• Tem Hit Rate alto num servidor físico com poucas

tarefas e poucos processadores• Tem Hit Rate próximo de zero num servidor físico com

milhares de tarefas e muitos processadores• Contra-indicado para virtualização ou ambientes de

múltiplos workloads

• Cache L3 “off-chip” (shared)• Tem Hit Rate altíssimo em qualquer situação, mas sua

latência é maior que a do Cache “on-chip”• É menos performático para workloads únicos ou em

pequena quantidade• IMBATÍVEL em ambientes virtualizados ou de múltiplos

workloads

z9 z10 z196 Power6 Power7 Neh Itanium Sparc VIICore's por Chip 2 4 4 2 8 4 4 4Clock (MHz) 1700 4400 5200 4700 4100 3200 1860 2520L1 (KB) por core 512 192 192 128 64 64 32 128L2 (KB) por core 0 3 1536 2048 256 256 1024 6L3 (MB) por chip 0 0 24 32 32 8 24 0

L4 (MB) off-chip (shared) 160 192 768 Zero Zero Zero Zero Zero

Compartilhado por 64 core'sCompartilhado por 74 core'sCompartilhado por 96 core's



CacheConclusão:

Hit Rate alto permite que o processador fique menos tempo esperando.

Seu GHz é mais útil !

(e por GHz, entenda-se kW/h, BTU, etc)



Ambiente típico atual de DataCenterToday… Many System z shops run their core, data-sensitive business processes on System z but also maintain distributed servers to accommodate a variety of processing that interacts with mainframe resources.

Data Management StructureCore Business Applications

ConsolidationSecurityLow TCO

System z

The mainframe has hosted most of the world’s business data and executed a majority of core business transactions

The mainframe is best suited for applications that require the highest levels of:

– Reliability– Security– Performance– Availability– Service– Scalability

However, the mainframe is not optimized for all workloads

CyclesMany ISV Applications

Large Skill PoolsLow TCA

Distributed Systems

There are particular industry application where distributedtopology is the standard.

Distributed environments are suited for applications where:– High levels resource sharing is not required– The workload is

• CPU intensive• Does not require high levels of data sharing• Highest levels of availability are not required

Other Distributed System Characteristics -– Multiple management interfaces– Redundancy required for availability– On-site assembly– End-user service for problem resolution



Server

Compute StorageMemory Network

Operating System

Software

Compute StorageMemory NetworkCompute StorageMemory Network

Operating System

Software

Virtualized Server

Virtualization

Virtual Server

OSSW

Virtual Server

OSSW

Virtual Server

OSSW

Compute StorageMemory Network

Virtualization

Virtual Server

OSSW

Virtual Server

OSSW

Virtual Server

OSSW

Virtual Server

OSSW

Virtual Server

OSSW

Virtual Server

OSSW

Compute StorageMemory NetworkCompute StorageMemory Network

Virtualized Cluster

Mobility

Optimized for ….• Availability• Performance• Energy

Mobility


Mobility


Virtualized Heterogeneous

Cluster

Advanced Management of Virtualized IT Infrastructure

t

The Evolution of the “Platform”


IBM zEnterprise

IBM System z

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 …

Linux em IBM System z

Uma idea simples que mudou o mundo !!!!



"Hello everybody... I'm doing a (free) operating system (just a hobby, won't be b ig and professional...)."

Linus Torvalds, crea tor o f L inux, from the first Internet announcement on August 25 , 1991. Even he initia lly underestimated its potentia l.



Inovações e Riscos

Will Linux be accepted by clients on the world’s most secure server platform?

– Free Software Foundation UNIX®-like operating system– GNU - GNU is Not UNIX– GPL - GNU General Public License

December 1999: IBM published a collection of patches/additions to the Linux 2.2.13 kernel for System/390® to start a market evaluation

1999



O Linux veio para ficar

Programadores: Trabalho por obrigação Versus trabalho por paixão

GPL é copyright reverso: propriedade da Humanidade

Independência de plataforma: – Não faz sentido portar uma aplicação Solaris para HP-UX, AIX ou Windows e continuar

preso a um hardware específico– Linux é alheio ao roadmap dos fabricantes: vai para onde o hardware for

Apoio de todos os fabricantes: aposta estratégica ou falta de opção

Mercado SMB de serviços: fomento sem precedentes



TCP/IPHTML

HTTP Browsers

Web Servers

GUIs

SSLSET

Java™

Open Standards

Open Source

LinuxXML

Inovação, Visão e Direção Estratégica

“Linux will do for applications,what the Internet did for networks”

source: Irving Wladawsky-Berger

2000



Free Software Foundationhttp://www.fsf.org



Os Valores do Linux e Software “Livre”

Industry Wide Initiativefirst operating system that the entire industry has rallied around. Not just

select vendors

Multi-platformruns on every platform. There is no

other OS with this characteristic

Growing Marketplace Acceptance

continual compound growth of deployments

Application mobilityLinux® makes it a reality. Increased

solution availability via Linux application portfolio

Enabling Business Innovation

based on open standards it is base for innovative uses of technology

Community Developed & Owned

collaboration for Linux among the best programmers. It is freely available

Large Skill Poolpreferred by university

students

LinuxLinux

"It's out there... LINUX is already inside most companies in some form or another and will continue to spread - with or without the CIO's support"

- CIO Magazine, 12/15/2000

1999



Linux em IBM System z? Não é um Linux especial

– Tudo relevante a Linux & System z foi oferecido para a comunidade de Open Source

– Aproximadamente 1% mudança de código– O Linux é executado diretamente no hardware

– Ambiente puro ASCII

– Utiliza dos recursos do Mainframe além do FCP– Não é um substitudo de nenhum outro sistema operacional em

IBM System z

Linux

Kernel

GNU

RTL

GNU

Binutils

Linux applications

GNU

Compiler

System z Hardware

IBM developed code

IBM developed code



Linux and z/VM Technology ExploitationGerenciamento Cooperativo de Memória (CMM)

Solução: falta de memória real detectada pelo z/VM Virtual Machine Resource Manager (VMRM)

Servidores Linux sinalizados para reduzir o consumo de memória

Páginas de memória do Linux são liberadas

A demanda por memória real e discos de paginação do z/VM é reduzida

Ajuda a melhorar a performance do ambiente como um todo e a performance individual dos servidores virtuais

VirtualMemory

RealMemory

LinuxLinuxLinuxLinuxLinux

Expanded Storage

Disk space

z/VM Paging Subsystem

= Inactive virtual memory = Active virtual memory

z/VM

Learn more at:ibm.com/servers/eserver/zseries/zvm/sysman/vmrm/vmrmcmm.html



Extreme Virtualization with Linux on z/VMLinux Exploitation of z/VM Virtual Disks in Storage (VDISK)

VirtualMemory

RealMemory

VDISK support is Data-in-Memorytechnology

– Simulate a disk device using real memory– Achieve memory speeds on disk I/O operations– VDISKs can be shared among virtual machines

Linux exploitation: high-speed swap device

– Use VDISKs for Linux swap devices instead of real disk volumes

– Reduces demand on I/O subsystem– Helps reduce the performance penalty normally

associated with swapping operations– An excellent configuration tool that helps clients

minimize the memory footprint required for virtual Linux servers

– Helps improve the efficiency of sharing real resources among virtual machines

Linux Linux Linux Linux Linux

VDISK VDISK VDISK VDISK VDISK



A Simple Idea

Increased solutions through Linux application portfolio Large number of highly skilled programmers familiar with Linux Integrated business solutions

– Data richness from IBM eServer™ zSeries®

– Web capability of Linux applications

Industrial strength environment– Flexibility and openness of Linux

– Qualities of service of zSeries and S/390®

Unique ability to easily consolidate large number of servers

2000



Uma Idéia Simples que mudou o mundoLinux on IBM System z® is providing a scalable and flexible virtual infrastructure inside a single IBM System z server.

It is customized for server virtualization and consolidation, business relevant workloads such as business intelligence and collaboration, and new business models such as cloud computing.

The Linux on System z infrastructure provides significant IT cost savings opportunities:

– People Cost – increase the productivity of the IT staff

– Software Costs – reduce software license fees by consolidating a large number of virtual machines per System z core

– Energy Costs – dramatic reduction in power usage

– Facilities Costs – reduce floor space with dense server consolidation on System z

– Networking Costs – consolidate networks inside the box too

2010



SuSE 7.0

SLES8

SLES11

SLES10 RHEL5

RHEL3RHEL4

& SLES9

SLES7

Linux, z/VM e System z – vantagens exclusivas com Inovação Tecnológica

z10 EC z10 BC

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 …

z900 z800 z990 z890

z9 EC z9 BC

G5 G6

z/VM V4.3

z/VM V4.4

z/VM V5.2

z/VM V5.3

z/VM V6.1

z/VM V5.1

z/VM V5.4

z/VM V4.1/V4.2



Unique Linux Extensions to Leverage Powerful System z Technology Advantages

SLES11

SLES10

RHEL5

RHEL3

RHEL4 & SLES9

HMC management of Linux guests IBM Director – for

Systems Mgmt.

Platform for System/390Kernel 2.2

Multipath I/OFCP / SCSIVirtual Server SupportVIPA Load Balancing

64-bit Exploitation Improved performance

of SCSI disk I/OEnhanced performance

assists for z/VM®

guestsMemory management

optimization (CMMA and VMRM-CMM)

Systems Management APIs to manage Linux virtual machinesSCSI Disks,

including guest IPL >16 CPs

HiperSockets™

Dynamic CPU & I/O enablementHigh availability

CPU & Memory Recovery SMP Scalability SAN enablementVLANKernel 2.4

RAS ImprovementsSysfsKernel 2.6

NPIVLarge Page supportCPU Node AffinityAccountingCrypto Dynamic Add

HyperSwap™ - Disk MirroringAuto-adaptive CPU

& Memory Mgmt.Shutdown ActionsCollaborative

Memory Mgmt.

Improved scalability and constraint reliefSupport for z/VM-mode

partitionsDynamic virtual

processor managementDump Linux guests to

SCSI disksDynamic storage

reconfigurationSupport for Extended

Address Volumes

VSWITCH FCP List Directed IPL

Guest Virtual Network

FCP support Improved I/O

performance for Linux guests Enhanced page fault

support for Linux guests

Dynamic memory attach / detachDecimal Floating PointHyper PAV enablemt.Kernel Message Cat.Vertical CPU Mgmt.

SLES8

SLES7

SuSE 7.0



A Simple Idea that evolved to an Uniquely Powerful Solution

Single-server simplicity– Saving opportunities in software and management costs, power and floor space– A cluster with one machine backing up another and an additional failover machine

Advanced resource utilization and dynamic allocation– Industry leading virtualization and sharing of system resources such as processors, memory,

communication, storage, I/O, networking

Massive scalability– Running up to thousands of virtual Linux servers concurrently– Supporting a broad range of solutions such as cloud computing, business intelligence,

collaboration and Web application serving

Rock-solid system security and reliability– Most secure commercial server1 ensuring isolation of each virtual server environment– Cost-attractive business resilience and failover solutions1 IBM System z servers are the world’s only servers with the highest level of hardware security certification,Common Criteria Evaluation Assurance Level 5 (EAL5).

Consolidate more & spend less



2001

HA ClusteringIMS CON.TSM 4.2

2000

VM/ESA 2.4 Kernel 2.2 DB2 UDB 7.1 DB2 CON. 7.1WAS 3.5

Pilot

Continuous Improvements and DeploymentsFrom the Beginning till Today – Real client case

2002

z/VM 4.1 Kernel 2.4 WAS 4.0 MQ Series

Web-mail Contract based on Usage

2003

z/VM 4.3DB2 UDB 8.1DB2 CON. 8.1TSM 5.1WAS 5.0ORACLE 9.0.1

Employee PortalE-Ticketing

2004

WAS 5.1ITSA 1.2TMTP 5.2ORACLE 9.2.0

Domestic Rev. AccountingInternational Refund SystemFinancial Information

z/VM 5.1TMWI, TMDB DB2 UDB 8.2(64bit based)

Frequent Travel Bonus

2005~2010

Total 54 Servers

32 Production

22 Development & Test

28 WebSphere Application

7 Oracle DB

5 DB2 Universal Database™

DNS

Web Reporting

XecureWeb

On Demand

Disaster Recovery

Resource Sharing thru Virtualization

Consolidation

Stability

Performance

Standardization

Revenue accounting systemTravel schedule inquiry



Transformação da IBM entrega resultados de sucesso Percebendo o Valor da Virtualização

80% less energy and 85% less floor space.2x existing capacity, no increase in consumption or impact by 2010

Hardware ConfiguratorsHardware Configurators

On Demand WorkplaceOn Demand Workplace

AHE Domino®, GNA, GWAAHE Domino®, GNA, GWA

Partner CommercePartner Commerce

SWG, IGS Internet SitesSWG, IGS Internet Sites

GWA eliminationGWA eliminationIntranet Labor ClaimingIntranet Labor Claiming

Integrated Demand SupplyIntegrated Demand SupplyPre-sales, Post-SalesPre-sales, Post-Sales

CMVC, AHE, Cognos®CMVC, AHE, Cognos®

Tivoli® TADDM, CCMDB, TSRM, Maximo®

Tivoli® TADDM, CCMDB, TSRM, Maximo®

e-Maile-Mail

2Q07 3Q07 4Q07 1Q08 3Q08 4Q08 3Q09

Cumulative Migrations

Financial Risk ManagementFinancial Risk Management

2Q08 4Q091Q09 2Q09



Campanha dos 10 anos do Linux em Mainframe



Benchmark nativo 64-bit versão de Lotus®

Domino® v8.5 no z10 EC mostrou :

– 102 mil usuários concorrentes– Cerca de1.5 million Domino

transações a cada 10 minWhile the z/VM virtualization environment had less than a cumulative 2% overhead for each Lotus Domino server instance, that benchmark is not typical of most other virtualization technologies.

Also, the low cost associated with the ability to virtualize massive amounts of I/Os can allow for significant vertical scalability and cost savings.

Understanding the cost of your virtual environment is critical to building a successful virtual deployment.

1000000

1200000

1400000

1600000

Total Domino Transaction10 Minute Intervals

Transactions

85000

90000

95000

100000

105000

Connected and Active 15 Minute Users10 Minute Intervals

Connected Active 15

Benchmark:

ibm.com/developerworks/lotus/library/domino-linux

ibm.com/systems/z/os/linux/solutions/ld.html

IBM Lotus Domino, Linux, Virtualização, Escalabilidade:


Entregando informação confiavel para analise de negócios mais inteligentes

Analyze

Integrate

Transactional & Collaborative Applications

Manage

Business Analytics Applications

External Information Sources

Data Cubes/MiningInfoSphere Warehouse

Streams InfoSphereStreams

Big DataInfoSphere BigInsights

Master Data

Content

Data

StreamingInformation Govern

QualitySecurity &

PrivacyLifecycle

Data Warehouses

Standards

Content IBM Content Analytics


Micro-Cluster:1

Micro-Cluster:2

Micro-Cluster:3

Micro-Cluster:N

PosteriorsPosteriors“Relevant Action”

Highest probability action bet accounting for Priors and current

Conditions.

Channel or Device ExecutionChannel or Device Execution

APIAPI

Value TiersValue Tiers

Complex Event ProcessingClustering & Segmentation

Decision TradeDecision Trade--offoff

(ActionEngine)Actions, messages, offers are pieced together based

on Posterior bet.

Bayesian Network

ConditionalConditional“Recognize Me”

How is the customer interacting right now?PriorsPriors

“Remember Me”How has the

customer engaged in the past?

Exist

ing

cust

omer

cla

ssifi

catio

n

Call

Cent

er

Inte

rnet

Mob

ile

Existing infrastructure we leverage

Objects we bring

Know

ledg

e da

taba

se w

e co

dify

API’s

we

deve

lop

Anal

ytic

s w

e cu

stom

ize

Decisions we model

Even

t pro

cess

ing

tool

s w

e br

ing

2011 - IBM Confidential & Proprietary

New

Lapsed

Uncom

mitted

Best

11%

24%

9%

31%

34%

New

Lapsed

Uncom

mitted Silver

Gold

11%

9%

31%

25%

Fluxo de decisão – qual o melhor próximo passo ao agir



O patrocinador do projeto necessita ter uma visão global

Investimentos Estratégicos serão necessários para a migração

Objetivos claros, times dedicados, inclusive lideres são necessários para executar a migração no prazo e investimentos projetados

Alavancar talentos e capacidades disponiveis na IBM para alcançar os resultados

Fatores Críticos de Sucesso

ZSP03050-USEN-00



DE PARA

Hardware Atual Sun and HP servers z10 EC

Espaço Físico 61 1

Core’s/Memória 442 core’s / 1440 GB 16 Cores IFL’s / 82GB

Utilização Média 13.3% 40%

Pico de Utilização 28.7% 92%

# DBs, size of DB 61 61

Aplicação Oracle databases Oracle databases

OS Sun Solaris / Linux + z/VM

Utilização de Energia:Power & cooling (Whr)Heat (BTUs/hr)

345,618 Whr737,030 BTUs/hr

14,766 Whr -> 95% less39,648 BTUs/hr -> 94% less

Resumo dos Benefícios: Economia de software e energia elétrica, Melhor utilização

Cliente do setor financeiro: 61 servidores Sun e HP em uma z10

Data is based on real client opportunity and on internal standardized costing tools and methodologies.Client results will vary by types of workloads, technology level of consolidated servers, utilization factor, and other implementation requirements. Savings will vary by client.

VIRTUALIZATION



Dados de Abril de 2.008 – sem atualizações posteriores



System z Technical University ibm.com/systems/conferenceseries

www.ibm.com/ibm/licensing/patents/pledgedpatents.pdf500 patentes para serem usadas com Open SourceIniciativas da IBM com Open Source



Perguntas ? ...



Obrigado !

João Paulo [email protected]


What is Watson?

A computing system named after our founder Thomas J. Watson.

Rivals a human’s ability to answer questions posed in natural language with speed, accuracy and confidence.

Leverages the power of IBM Content Analytics to perform lexical analysis as part of natural language processing.

Harnesses a cluster of IBM’s commercially-available POWER7 system optimized to process thousands of simultaneous tasks at rapid speeds.

Incorporates a number of IBM exclusive technologies for the specialized demands of processing an enormous number of concurrent tasks and data while analyzing information in real time.

This is not search, where a query retrieves a list of links to potential answers based on popularity and page ranking.

Watson’s ability to understand the meaning and context of human language, and rapidly process information to find precise answers to complex questions, holds enormous potential to transform how computers can help people accomplish tasks in business and their personal lives.


Watson’s use of IBM Content Analytics

Watson leverages the power of natural language processing as the cornerstone to translate interactions between computers and human (natural) languages.

A critical step in natural language processing is lexical analysis - the process of converting a sequence of characters into a sequence of tokens, the key step in making text understandable (or computable).

A feature of IBM Content Analytics (known as LanguageWare) is performing the Lexical Analysis function in Watson as part of natural language processing.

IBM Content Analytics is a platform to derive rapid insight. Transform raw information into business insight quickly without building models or deploying complex systems. Derive insight in hours or days … not weeks or months.

External and Internal Content (and Data) Sourcesincluding Social Media and More

Rapidly Derived Insight

Analyzeand Visualize

Searchand Explore

Aggregateand Extract

Lexical Analysis

IBM Content Analytics

NTT DoCoMo leverages IBM Content Analytics to improve service and reduce customer churn

“Insight into customer interaction logs is an information gold mine for us.”

— General Manager Japan Business


Taking on Jeopardy! – A Grand Challenge

On February 14, Watson will challenge Jeopardy! world champions Ken Jennings and Brad Rutter in a two-match contest to be aired over three consecutive nights.

Jeopardy! - an American quiz show - covers a broad range of topics, such as history, literature, politics, arts and entertainment, and science.

Jeopardy! poses a grand challenge for a computing system:

– Broad range of subject matter– Speed of accurate responses and confidence– Requires analyzing subtle meaning, irony, riddles,

and other complexities.

Beyond Jeopardy!, the technology behind Watson can be adapted to solve business and societal problems – for example, diagnosing disease, handling online technical support questions, and parsing vast tracts of legal documents - and to drive progress across many industries.


A aplicabilidade dos conceitos do Computador Watson Watson provides us with an opportunity to differentiate IBM today and put into perspective for clients that:

Systems Design Matters– Watson demonstrates the future of computing systems design – how software and hardware

is architected, integrated and optimized to solve business problems. – We call this workload optimized systems. This is how Watson was built. – Watson harnesses IBM’s commercially available, workload optimized system which can

process thousands of simultaneous tasks at rapid speeds.

Data and Content Matter– In today’s era of massively increasing amounts of information, data is becoming an

organization's most important asset • ‘Data’ includes both structured data in databases, and unstructured data (or content)

– Watson leverages capabilities found in IBM Content Analytics to understand text as part of natural language processing, and shares “Big Data” management technology found in InfoSphere software

– Watson demonstrates how organizations can apply analytics to gain meaningful insights from information, anticipate and shape better business outcomes, and improve their business operations and services to their customers.

There is a Smarter Way of Getting Things Done– Watson demonstrates how industries and businesses can be transformed through the

application of new technology advances (Smarter Planet).– Beyond Jeopardy!, Watson-like capabilities applied to healthcare, government, transportation,

and other industries can enable Smarter Planet transformations.


Watson demonstrates the power of analytics and the ability to gain meaningful and actionable insights from an ever-increasing volume and variety of data - in real time.

Analytics can help organizations anticipate and shape better business outcomes, and improve their business operations and services to their customers.

Watson-like analytics capabilities are available and driving business outcomes for our client’s today, including:

– Bharti AirTel (India) – Applying analytics allows Bharti's retailers and call center agents to provide the right offer to the right customer, and to pinpoint where to make infrastructure investments based on market demographics and growth patterns.

– The New York State Department of Taxation and Finance – Using analytics to transform its approach from “pay and chase” to “next best case”. The system identifies the next refund requests most likely to be questionable and focuses audit resources on these. In its five years of operation, the system has preserved more than $889 million against fraudulent requests.

The Future of Data: Business Analytics Top-performing companies are three times more likely to be leading users of analytics, according to a 2010 report by the IBM Institute for Business Value and MIT Sloan Management Review.

Education

Convergências, Integração e Desempenho no IBM System z Enterprise