IBM Netezza TwinFin® Líder em Appliances para Data Warehouse

Apresentação em tema: "IBM Netezza TwinFin® Líder em Appliances para Data Warehouse"— Transcrição da apresentação:

1 IBM Netezza TwinFin® Líder em Appliances para Data Warehouse
Silvio Ferrari IBM Netezza Systems Engineer

2 Netezza, IM e BAO Analisar Integrar Gerenciar Governança Integrate &
Aplicações Transacionais & Colaborativas Aplicações Analíticas de Negócio Integrar Analisar Master Data Big Data Data Warehouses www Gerenciar Dados Estruturados Integrate & Cleanses Fontes de informação Externas Data Warehouse Appliances Dados Streams Conteúdo Informação Streaming Segurança & Privacidade Governança Gerenciamento de Lifecycle Qualidade 2 2 2

3 Verdadeiros Appliances
Dispositivos especializados Otimizados para um propósito Solução completa Instalação rápida Operação muito simples Interfaces padrão de mercado Baixo custo Before we get into what Netezza appliances are, let’s agree on what appliances are in general Appliances—ie black-box solutions—are commonplace in the IT industry We are all familiar with how they have simplified operations and revolutionized entire markets With networking appliances, even the most incompetents when it comes to IT can create a multi-user network in their homes in a matter of minutes The iPod is a great example of an appliance that simplified and revolutionized digital entertainment instead of using a PC to do the same function These appliances have some common attributes that make them very attractive compared to the old way of doing things They do only one thing, but do it better than any alternative They are plug-and-play, with a simple interface that anyone can operate And they are generally much cheaper than the alternative Netezza anuncia servidor em 2002 Está no melhor quadrante do Gartner desde 2008 2008 Data Warehouse Database Management Systems Magic Quadrant report released on December 23, 2008

4 A Simplicidade de um Appliance
Netezza

5 Carregando dados no Appliance IBM Netezza
Integração de dados Ab Initio Business Objects/SAP Composite Software Expressor Software GoldenGate Software (Oracle) Informatica IBM Information Server Sunopsis (Oracle) WisdomForce ... e outras mais.... Inserindo SQL ODBC JDBC OLE-DB Constant 2TB per hour loads with little adverse impact on queries. Opportunity: move from overnight batch loads to trickle-feeds as events occur. Wide range of complementary vendors For Oracle customers – move from row-based PL/SQL to set-based Extract – Load – Transform in-place of ETL. AOL example replace Sun / Oracle ETL stages Saves more than 7 million dollars a year Data integration partners support this “push-down processing” technique.

6 Consultando o Appliance IBM Netezza
Reporting e Análise Actuate Business Objects/SAP Cognos (IBM) Information Builders Kalido KXEN MicroStrategy Oracle OBIEE QlikTech Quest Software SAS SPSS (IBM) Unica (IBM) ... e outras mais.... extraindo SQL ODBC JDBC OLE-DB Once loaded – data is available There are no indexes and aggregates to update before data can be queried Partnerships with all major BI vendors While SQL-based reports are common analytics using tools such as SAS and SPSS derive greater value from the same data. I’ll now investigate TwinFin’s in-database analytics

7 A arquitetura IBM Netezza AMPP™ ( parte de Hardware )
FPGA CPU Analíticos Avançados Memory Hosts Host BI FPGA CPU Memory ETL FPGA CPU Loader Memory Rede Interna Discos Applicações S-Blades™ Netezza Appliance 7 7

8 Servidores Blade Memória CPUs 8 8

9 Acelerador IBM Netezza Database
Memória CPUs FPGA 9 9

10 Elimina colunas não usadas Operações complexas: ∑
Nosso segredo: select DISTRICT, PRODUCTGRP, sum(NRX) from MTHLY_RX_TERR_DATA where MONTH = ' ' and MARKET = and SPECIALTY = 'GASTRO' FPGA CPU A key component of Netezza’s performance is the way in which its streaming architecture processes data. The Netezza architecture uniquely uses the FPGA as a turbocharger … a huge performance accelerator that not only allows the system to keep up with the data stream, but it actually accelerates the data stream through compression before processing it at line rates, ensuring no bottlenecks in the IO path. You can think of the way that data streaming works in the Netezza as similar to an assembly line. The Netezza assembly line has various stages in the FPGA and CPU cores. Each of these stages, along with the disk and network, operate concurrently, processing different chunks of the data stream at any given point in time. The concurrency within each data stream further increases performance relative to other architectures. Compressed data gets streamed from disk onto the assembly line at the fastest rate that the physics of the disk would allow. The data could also be cached, in which case it gets served right from memory instead of disk. The first stage in the assembly line, the Compress Engine within the FPGA core, picks up the data block and uncompresses it at wire speed, instantly transforming each block on disk into 4-8 blocks in memory. The result is a significant speedup of the slowest component in any data warehouse—the disk. The disk block is then passed on to the Project engine or stage, which filters out columns based on parameters specified in the SELECT clause of the SQL query being processed. The assembly line then moves the data block to the Restrict engine, which strips off rows that are not necessary to process the query, based on restrictions specified in the WHERE clause. The Visibility engine also feeds in additional parameters to the Restrict engine, to filter out rows that should not be “seen” by a query e.g. rows belonging to a transaction that is not committed yet. The Visibility engine is critical in maintaining ACID (Atomicity, Consistency, Isolation and Durability) compliance at streaming speeds in the Netezza. The processor core picks up the uncompressed, filtered data block and performs fundamental database operations such as sorts, joins and aggregations on it. It also applies complex algorithms that are embedded in the snippet code for advanced analytics processing. It finally assembles all the intermediate results together from the entire data stream and produces a result for the snippet. The result is then sent over the network fabric to other S-Blades or the host, as directed by the snippet code. Elimina colunas não usadas Operações complexas: ∑ Joins, Aggs, etc. Restringe Visibilidade Parte da tabela MTHLY_RX_TERR_DATA (comprimida) Descomprime select DISTRICT, PRODUCTGRP, sum(NRX) where MONTH = ' ' and MARKET = and SPECIALTY = 'GASTRO' sum(NRX) 10

11 O S-Blade™ IBM Netezza 11 11

12 Arquitetura IBM Netezza TwinFin™
Hardware+Software Otimizados Projetado (e não simplesmente adaptado) para tarefas analíticas de alta performance; Não necessita ajustes; Dados Streaming Aceleradores de query por Hardware, para resultados mais rápidos Analíticos avançados Analíticos complexos executados in-database Verdadeiro MPP Todos os processadores totalmente utilizados para máxima eficiência e velocidade 1212 12

13 Simplicidade do Appliance IBM Netezza ( Software )
Sem índices ou ajustes DBAs se tornam Gerenciadores de Dados, em vez de administradores de banco de dados Administração de storage desnecessária dbspace/tablespace: não há sizing ou configuração redo/physical/Logical log: não há sizing ou configuração page/block de tabelas: não há sizing ou configuração extent para tabelas não há sizing ou configuração Temp Space: não há alocação ou monitoração dbspaces: não há decisões para nível RAID Logical Volume: não há criação de files OS kernel: não há alterações OS kernel: não há níveis de patch requeridos Sessões JAD para configurar host/network/storage não requeridas We do not have indexes. They are not an option, they simply do not exist. There is no disk administration or SA administraion. Day 2, the customer has a pool of disk performant ready. Upgrades are performed by Netezza as standard maintenance tech support call. Does Oracle help you go from 9i to 10g? Instead of spending time and effort on tedious DBA tasks, use the time for higher BUSINESS VALUE tasks: Bring on new applications and groups Quickly build out new data marts Provide more functionality to your end users Passos da instalação: - conectar energia elétrica - rodar testes (8h) - entregar servidor ao cliente Sem instalação de software

14 Complexidade versus Simplicidade IBM Netezza Criando um database:
0. CREATE DATABASE TEST LOGFILE 'E:\OraData\TEST\LOG1TEST.ORA' SIZE 2M, 'E:\OraData\TEST\LOG2TEST.ORA' SIZE 2M, 'E:\OraData\TEST\LOG3TEST.ORA' SIZE 2M, 'E:\OraData\TEST\LOG4TEST.ORA' SIZE 2M, 'E:\OraData\TEST\LOG5TEST.ORA' SIZE 2M EXTENT MANAGEMENT LOCAL MAXDATAFILES 100 DATAFILE 'E:\OraData\TEST\SYS1TEST.ORA' SIZE 50 M DEFAULT TEMPORARY TABLESPACE temp TEMPFILE 'E:\OraData\TEST\TEMP.ORA' SIZE 50 M UNDO TABLESPACE undo DATAFILE 'E:\OraData\TEST\UNDO.ORA' SIZE 50 M NOARCHIVELOG CHARACTER SET WE8ISO8859P1; 1. Oracle* table and indexes   2. Oracle tablespace     3. Oracle datafile       4. Veritas file         5. Veritas file system            6. Veritas striped logical volume               7. Veritas mirror/plex                 8. Veritas sub-disk                   9. SunOS raw device                      10. Brocade SAN switch                        11. EMC Symmetrix volume                          12. EMC Symmetrix striped meta-volume                             13. EMC Symmetrix hyper-volume                                 14. EMC Symmetrix remote volume (replication)                                 15. Days/weeks of planning meetings Mudar pata 6data!!!!!!! IBM Netezza: ZERO parâmetros: CREATE DATABASE my_db; Traditional architectures are much more compicated then just Stoarge + HW + RDBMS. There are multiple hops for the data. Mutliple areas of tuning. Either the customer does this themselves or pays someone to do it. 14 14

15 Simplicidade Netezza: criando uma tabela
ORACLE CREATE TABLE "MRDWDDM"."RDWF_DDM_ROOMS_SOLD" ("ID_PROPERTY" NUMBER(5, 0) NOT NULL ENABLE, "ID_DATE_STAY" NUMBER(5, 0) NOT NULL ENABLE, "CD_ROOM_POOL" CHAR(4) NOT NULL ENABLE, "CD_RATE_PGM" CHAR(4) NOT NULL ENABLE, "CD_RATE_TYPE" CHAR(1) NOT NULL ENABLE, "CD_MARKET_SEGMENT" CHAR(2) NOT NULL ENABLE, "ID_CONFO_NUM_ORIG" NUMBER(9, 0) NOT NULL ENABLE, "ID_CONFO_NUM_CUR" NUMBER(9, 0) NOT NULL ENABLE, "ID_DATE_CREATE" NUMBER(5, 0) NOT NULL ENABLE, "ID_DATE_ARRIVAL" NUMBER(5, 0) NOT NULL ENABLE, "ID_DATE_DEPART" NUMBER(5, 0) NOT NULL ENABLE, "QY_ROOMS" NUMBER(5, 0) NOT NULL ENABLE, "CU_REV_PROJ_NET_LOCAL" NUMBER(21, 3) NOT NULL ENABLE, "CU_REV_PROJ_NET_USD" NUMBER(21, 3) NOT NULL ENABLE, "QY_DAYS_STAY_CUR" NUMBER(3, 0) NOT NULL ENABLE, "CD_BOOK_SOURCE" CHAR(1) NOT NULL ENABLE) PCTFREE 5 PCTUSED 95 INITRANS 4 MAXTRANS 255 STORAGE( FREELISTS 6) TABLESPACE "DDM_ROOMS_SOLD_DATA" NOLOGGING PARTITION BY RANGE ("ID_PROPERTY" ) (PARTITION "PART1" VALUES LESS THAN (600) PCTFREE 5 PCTUSED 95 INITRANS 4 MAXTRANS 255 STORAGE(INITIAL FREELISTS 6 FREELIST GROUPS 1) TABLESPACE "DDM_ROOMS_SOLD_DATA" NOLOGGING NOCOMPRESS, PARTITION "PART2" VALUES LESS THAN (1200) PCTFREE 5 PCTUSED 95 INITRANS 4 MAXTRANS 255 "DDM_ROOMS_SOLD_DATA" NOLOGGING NOCOMPRESS, PARTITION "PART3" VALUES LESS THAN (1800) PCTFREE 5 PCTUSED 95 INITRANS 4 MAXTRANS 255 "DDM_ROOMS_SOLD_DATA" NOLOGGING NOCOMPRESS, PARTITION "PART4" VALUES LESS THAN (2400) PCTFREE 5 PCTUSED 95 INITRANS 4 MAXTRANS 255 "DDM_ROOMS_SOLD_DATA" NOLOGGING NOCOMPRESS, PARTITION "PART5" VALUES LESS THAN (3000) PCTFREE 5 PCTUSED 95 INITRANS 4 MAXTRANS 255 "DDM_ROOMS_SOLD_DATA" NOLOGGING NOCOMPRESS, PARTITION "PART6" VALUES LESS THAN (MAXVALUE) PCTFREE 5 PCTUSED 95 INITRANS 4 MAXTRANS 255 "DDM_ROOMS_SOLD_DATA" NOLOGGING NOCOMPRESS ) ; Netezza CREATE TABLE MRDWDDM.RDWF_DDM_ROOMS_SOLD ( ID_PROPERTY numeric(5, 0) NOT NULL , ID_DATE_STAY integer NOT NULL , CD_ROOM_POOL CHAR(4) NOT NULL , CD_RATE_PGM CHAR(4) NOT NULL , CD_RATE_TYPE CHAR(1) NOT NULL , CD_MARKET_SEGMENT CHAR(2) NOT NULL , ID_CONFO_NUM_ORIG integer NOT NULL , ID_CONFO_NUM_CUR integer NOT NULL , ID_DATE_CREATE integer NOT NULL , ID_DATE_ARRIVAL integer NOT NULL , ID_DATE_DEPART integer NOT NULL , QY_ROOMS integer NOT NULL , CU_REV_PROJ_NET_LOCAL numeric(21, 3) NOT NULL , CU_REV_PROJ_NET_USD numeric(21, 3) NOT NULL , QY_DAYS_STAY_CUR smallint NOT NULL , CD_BOOK_SOURCE CHAR(1) NOT NULL) distribute on random; ORACLE Indexes CREATE INDEX "MRDWDDM"."RDWF_DDM_ROOMS_SOLD_IDX1" ON "RDWF_DDM_ROOMS_SOLD" ("ID_PROPERTY" , "ID_DATE_STAY" , "CD_ROOM_POOL" , "CD_RATE_PGM" , "CD_RATE_TYPE" , "CD_MARKET_SEGMENT" ) PCTFREE 10 INITRANS 6 MAXTRANS 255 STORAGE( FREELISTS 10) TABLESPACE "DDM_DATAMART_INDEX_L" NOLOGGING PARALLEL ( DEGREE 4 INSTANCES 1) LOCAL(PARTITION "PART1" PCTFREE 10 INITRANS 6 MAXTRANS 255 STORAGE(INITIAL NEXT MINEXTENTS 1 MAXEXTENTS PCTINCREASE 0 FREELISTS 10 FREELIST GROUPS 1 BUFFER_POOL DEFAULT) TABLESPACE "DDM_DATAMART_INDEX_L" NOLOGGING, PARTITION "PART2" PCTFREE 10 INITRANS 6 MAXTRANS 255 STORAGE(INITIAL NEXT MINEXTENTS 1 MAXEXTENTS PCTINCREASE 0 FREELISTS 10 FREELIST GROUPS 1 BUFFER_POOL DEFAULT) TABLESPACE "DDM_DATAMART_INDEX_L" NOLOGGING, PARTITION "PART3" PCTFREE 10 INITRANS 6 MAXTRANS 255 STORAGE(INITIAL NEXT MINEXTENTS 1 MAXEXTENTS PCTINCREASE 0 FREELISTS 10 FREELIST GROUPS 1 BUFFER_POOL DEFAULT) TABLESPACE "DDM_DATAMART_INDEX_L" NOLOGGING, PARTITION "PART4" PCTFREE 10 INITRANS 6 MAXTRANS 255 STORAGE(INITIAL NEXT MINEXTENTS 1 MAXEXTENTS PCTINCREASE 0 FREELISTS 10 FREELIST GROUPS 1 BUFFER_POOL DEFAULT) TABLESPACE "DDM_DATAMART_INDEX_L" NOLOGGING, PARTITION "PART5" PCTFREE 10 DEFAULT) TABLESPACE "DDM_DATAMART_INDEX_L" NOLOGGING, PARTITION "PART6" 1 BUFFER_POOL DEFAULT) TABLESPACE "DDM_DATAMART_INDEX_L" NOLOGGING ) ; ORACLE Bitmap index CREATE BITMAP INDEX "CRDBO"."SNAPSHOT_MONTH_IDX13" ON "SNAPSHOT_OPPTY_MONTH_HIST" ("SNAPSHOT_YEAR" ) PCTFREE 10 INITRANS 2 MAXTRANS 255 STORAGE(INITIAL NEXT MINEXTENTS 2 MAXEXTENTS PCTINCREASE 0 FREELISTS 1 FREELIST GROUPS 1 BUFFER_POOL DEFAULT) TABLESPACE "SFA_DATAMART_INDEX" NOLOGGING ; ORACLE Table Clusters CREATE CLUSTER "MRDW"."CT_INTRMDRY_CAL" ("ID_YEAR_CAL" NUMBER(4, 0), "ID_MONTH_CAL" NUMBER(2, 0), "ID_PROPERTY" NUMBER(5, 0)) SIZE 16384 PCTFREE 10 PCTUSED 90 INITRANS 3 MAXTRANS 255 STORAGE(INITIAL NEXT MINEXTENTS 1 MAXEXTENTS 1017 PCTINCREASE 0 FREELISTS 4 FREELIST GROUPS 1 BUFFER_POOL RECYCLE) TABLESPACE "TSS_FACT" ; As data volumes grow, oracle complexity increases. As new indexes are created in oracle, you break existing reports. All of this (indexes, partitioing) is an attempt to out guess the user’s data access. Netezza is database This is as complicated as it gets. Sem indexes Sem Admininstração ou ajustes Distribua os dados aleatoriamente, ou por Colunas 15 15

16 Complexidade Tradicional versus a Simplicidade Netezza (RDBMS 101)
CREATE TABLE EDW_PROD.EDW_RESPD_EXPSR_MIN_FACT ( RPT_PERIOD_DIM_ID NUMBER NOT NULL, SRVY_WEEK_DIM_ID NUMBER NOT NULL, DATE_DIM_ID NUMBER NOT NULL, SRVC_MKT_SEG_DIM_ID NUMBER NOT NULL, RESPD_HHLD_DIM_ID NUMBER NOT NULL, MDOTLT_DIM_ID NUMBER NOT NULL, LSTN_LOC_DIM_ID NUMBER NOT NULL, EXPSR_MIN_CNT NUMBER NOT NULL, RESPD_WGHT_NMBR NUMBER, PRELIM_DAILY_WGHT_NMBR NUMBER, FINAL_DAILY_WGHT_NMBR NUMBER, TIMESHIFT_SECOND_CNT NUMBER, BGN_EXPSR_UTC_TS DATE, END_EXPSR_UTC_TS DATE, BGN_EXPSR_LOCAL_TS DATE, END_EXPSR_LOCAL_TS DATE, BGN_BCST_UTC_TS DATE, END_BCST_UTC_TS DATE, BGN_BCST_LOCAL_TS DATE, END_BCST_LOCAL_TS DATE, SOURCE_ID VARCHAR2(50 BYTE), ACTIVE_IND CHAR(1 BYTE) DEFAULT 'Y‘ NOT NULL, INSERT_TS DATE NOT NULL, UPDATE_TS DATE NOT NULL, METADATA_ID NUMBER, MEDIA_CODE VARCHAR2(10 BYTE), MDOTLT_HIER_DIM_ID NUMBER, OUT_OF_MKT_IND CHAR(1 BYTE) ) 516 BASE TABLE PARTITIONS… TABLESPACE AT_EDW_REXMIN PCTUSED 0 PCTFREE 10 INITRANS 1 MAXTRANS 255 LOGGING PARTITION BY RANGE (RPT_PERIOD_DIM_ID) ( PARTITION RP0000 VALUES LESS THAN (0) NOLOGGING NOCOMPRESS STORAGE ( INITIAL K NEXT K MINEXTENTS MAXEXTENTS UNLIMITED PCTINCREASE BUFFER_POOL DEFAULT ), PARTITION RP0001 VALUES LESS THAN (2) PARTITION RP0002 VALUES LESS THAN (3) BUFFER_POOL DEFAULT ), … … PLUS DDL FOR 513 MORE PARTITIONS CREATE TABLE EDW_PROD.EDW_RESPD_EXPSR_MIN_FACT ( RPT_PERIOD_DIM_ID INTEGER NOT NULL, SRVY_WEEK_DIM_ID INTEGER NOT NULL, DATE_DIM_ID INTEGER NOT NULL, SRVC_MKT_SEG_DIM_ID INTEGER NOT NULL, RESPD_HHLD_DIM_ID INTEGER NOT NULL, MDOTLT_DIM_ID INTEGER NOT NULL, LSTN_LOC_DIM_ID INTEGER NOT NULL, EXPSR_MIN_CNT NUMERIC(9,2) NOT NULL, RESPD_WGHT_NMBR NUMERIC(9,2), PRELIM_DAILY_WGHT_NMBR NUMERIC(9,2), FINAL_DAILY_WGHT_NMBR NUMERIC(9,2), TIMESHIFT_SECOND_CNT INTEGER, BGN_EXPSR_UTC_TS TIMESTAMP, END_EXPSR_UTC_TS TIMESTAMP, BGN_EXPSR_LOCAL_TS TIMESTAMP, END_EXPSR_LOCAL_TS TIMESTAMP, BGN_BCST_UTC_TS TIMESTAMP, END_BCST_UTC_TS TIMESTAMP, BGN_BCST_LOCAL_TS TIMESTAMP, END_BCST_LOCAL_TS TIMESTAMP, SOURCE_ID VARCHAR(50), ACTIVE_IND CHAR(1) DEFAULT 'Y‘ NOT NULL, INSERT_TS TIMESTAMP NOT NULL, UPDATE_TS TIMESTAMP NOT NULL, METADATA_ID INTEGER, MEDIA_CODE VARCHAR(10), MDOTLT_HIER_DIM_ID INTEGER, OUT_OF_MKT_IND CHAR(1) ) distribute on random; Index REXMIN_SOURCE_ID_I on 515 PARTITIONS… CREATE INDEX EDW_PROD.REXMIN_SOURCE_ID_I ON EDW_PROD.EDW_RESPD_EXPSR_MIN_FACT (SOURCE_ID) TABLESPACE AI_EDW_REXMIN INITRANS 2 MAXTRANS 255 LOGGING LOCAL ( PARTITION RP0000 NOLOGGING NOCOMPRESS PCTFREE 10 STORAGE ( INITIAL K NEXT K MINEXTENTS MAXEXTENTS UNLIMITED PCTINCREASE BUFFER_POOL DEFAULT ), PARTITION RP0001 PARTITION RP0002 BUFFER_POOL DEFAULT ), … … PLUS DDL FOR 512 MORE PARTITIONS Oracle: 34,500 KB de DDLs Netezza: KB de DDLs Index REXMIN_LLOC_FK_BI on 515 PARTITIONS… CREATE BITMAP INDEX EDW_PROD.REXMIN_LLOC_FK_BI ON EDW_PROD.EDW_RESPD_EXPSR_MIN_FACT (LSTN_LOC_DIM_ID) TABLESPACE AI_EDW_REXMIN INITRANS 2 MAXTRANS 255 LOGGING LOCAL ( PARTITION RP0000 NOLOGGING PCTFREE 10 STORAGE ( INITIAL K NEXT K MINEXTENTS MAXEXTENTS UNLIMITED PCTINCREASE BUFFER_POOL DEFAULT ), PARTITION RP0001 ), … … PLUS DDL FOR 513 MORE PARTITIONS Index REXMIN_REHH_FK_BI on 515 PARTITIONS… CREATE BITMAP INDEX EDW_PROD.REXMIN_REHH_FK_BI ON EDW_PROD.EDW_RESPD_EXPSR_MIN_FACT (RESPD_HHLD_DIM_ID) TABLESPACE AI_EDW_REXMIN INITRANS 2 MAXTRANS 255 LOGGING LOCAL ( PARTITION RP0000 NOLOGGING PCTFREE 10 STORAGE ( INITIAL K NEXT K MINEXTENTS MAXEXTENTS UNLIMITED PCTINCREASE BUFFER_POOL DEFAULT ), PARTITION RP0001 ), … … PLUS DDL FOR 513 MORE PARTITIONS Index REXMIN_SMS_FK_BI on 515 PARTITIONS… CREATE BITMAP INDEX EDW_PROD.REXMIN_SMS_FK_BI ON EDW_PROD.EDW_RESPD_EXPSR_MIN_FACT (SRVC_MKT_SEG_DIM_ID) TABLESPACE AI_EDW_REXMIN INITRANS 2 MAXTRANS 255 LOGGING LOCAL ( PARTITION RP0000 NOLOGGING PCTFREE 10 STORAGE ( INITIAL K NEXT K MINEXTENTS MAXEXTENTS UNLIMITED PCTINCREASE BUFFER_POOL DEFAULT ), … … PLUS DDL FOR 514 MORE PARTITIONS Index REXMIN_SRWK_FK_BI on 515 PARTITIONS… CREATE BITMAP INDEX EDW_PROD.REXMIN_SRWK_FK_BI ON EDW_PROD.EDW_RESPD_EXPSR_MIN_FACT (SRVY_WEEK_DIM_ID) TABLESPACE AI_EDW_REXMIN INITRANS 2 MAXTRANS 255 LOGGING LOCAL ( PARTITION RP0000 NOLOGGING PCTFREE 10 STORAGE ( INITIAL K NEXT K MINEXTENTS MAXEXTENTS UNLIMITED PCTINCREASE BUFFER_POOL DEFAULT ), … … PLUS DDL FOR 514 MORE PARTITIONS Index REXMIN_RP_FK_BI on 515 PARTITIONS… CREATE BITMAP INDEX EDW_PROD.REXMIN_SRWK_FK_BI ON EDW_PROD.EDW_RESPD_EXPSR_MIN_FACT (SRVY_WEEK_DIM_ID) TABLESPACE AI_EDW_REXMIN INITRANS 2 MAXTRANS 255 LOGGING LOCAL ( … … PLUS DDL FOR 515 PARTITIONS Index REXMIN_DATE_FK_BI on 515 PARTITIONS… CREATE BITMAP INDEX EDW_PROD.REXMIN_DATE_FK_BI ON EDW_PROD.EDW_RESPD_EXPSR_MIN_FACT (DATE_DIM_ID) TABLESPACE AI_EDW_REXMIN INITRANS 2 MAXTRANS 255 LOGGING LOCAL ( … … PLUS DDL FOR 515 PARTITIONS Index REXMIN_MEDO_FK_BI on 515 PARTITIONS… CREATE BITMAP INDEX EDW_PROD.REXMIN_MEDO_FK_BI ON EDW_PROD.EDW_RESPD_EXPSR_MIN_FACT (MDOTLT_DIM_ID)… … PLUS DDL FOR TABLESPACE PARTITIONS 16

17 Comparação de requerimentos de redes (internas e externas)
Exadata (full rack) TwinFin12 (full rack) 22 IP addresses for the InfiniBand network - 68 IP addresses for Ethernet (for a single cluster) 5 IP addresses 10 network drops minimum (with 50+ reported as being typical 4 network drops Total: 90 endereços IP Total: endereços IP 17

18 Monitorando a distribuição dos dados com NzAdmin
Uma má distribuição. O usuário escolheu a(s) coluna(s) errada(s) para a distribuição dos dados. Nota: Neste caso, o usuário escolheu a primeira coluna da tabela como a coluna de distrubuição. Uma decisão incorreta. Updates: 10/29/03: J. Feinsmith – System no longer chooses the first column by default. 18

19 Uma boa Distribuição: 2.2 Trilhões de Registros
19

20 Monitoração: Distribuição homogênea dos dados no sistema
Análise de SKEW com relação ao sistema Deve haver uma carga de utilização equivalente entre as SPUs 20

21 Backup e Restore Integração e certificação com ferramentas líderes de mercado: Simplifica integração com as principais ferramentas de backup e restore Suporte a X/Open Backup Services API (XBSA) Certificação IBM Tivoli Storage Manager (TSM) Certificação Veritas NetBackup™ da Symantec Backup and Restore Incremental Diminui significativamente os tempos de backup comparados ao backup Full Disponível no utilitário NZBACKUP Restores tipo Full ou parcial Dom Seg Ter Qua Qui Sex Sab Full Dif Cumulativo 21

22 The IBM Netezza TwinFin™ - Expansão
Em caso de expansão: - um novo sistema completo é enviado - dados migrados ONLINE - IPs são redirecionados - servidor original é desligado e devolvido 2222 22

23 i-Class: Analytics Without Constraints
Big Data Big Math Analyze wider and deeper data Additional dimensions Richer history Increase computational intensity More complex models Faster execution for results 23

24 Advanced Analytics with TwinFin i-Class
SAS, SPSS Demand Forecasting SQL Fraud Detection R, S+ SQL 24

25 Simples de Instalar e Operar
Operações Simplesmente carregue e use… é um appliance! Instalação em ~2 dias! Fácil de avaliar e funciona como anunciado! Desenvolvedores BI & DBAs – mais ágeis Sem configuração ou modelagem física Sem índices ou ajustes – performance imediata Agnóstico a modelos de dados Data Architects / DBA focam nos negócios, não na modelagem física Desenvolvedores ETL Tabelas de agregação não necessárias – lógica de ETL simplificada Cargas e transformações mais rápidas Analistas de Negócio Análise “Linha de Pensamento”– 10 a 100x mais rápida Consultas ad hoc – sem ajustes, sem índices Consultas complexas a grandes datasets Menor latencia – cargas e consultas simultâneas processamento OnStream a centenas de nodes

26 Família de Appliances para todo o ciclo de gerenciamento:
Skimmer Sistemas de Desenvolvimento e Testes 1 TB to 10 TB TwinFin Data Warehouse Analítico de alta Performance 1 TB to 1.5 PB Cruiser Archiving acessível por SQL, Back-up / DR 100 TB to 10 PB

27 15,000 users running 800,000+ queries per day 50X faster than before
Speed 15,000 users running 800,000+ queries per day 50X faster than before “…when something took 24 hours I could only do so much with it, but when something takes 10 seconds, I may be able to completely rethink the business process…” - SVP Application Development, Nielsen Predictability Source: 27

28 Up and running 6 months before having any training
Simplicity Up and running 6 months before having any training DAYS WEEKS MONTHS 200X faster than Oracle system ROI in less than 3 months “Allowing the business users access to the Netezza box was what sold it.” Steve Taff, Executive Dir. of IT Services XO Communications offers a variety of communications services including voice over internet protocol (VoIP), data and internet services, network transport, broadband wireless access, and hosted and managed services. Its high capacity IP network and advanced transport network support more than 50 percent of the Fortune 500 and many of the world’s largest telecommunications companies. 28

29 7 years of historical data 100-200% annual data growth
Scalability 1 PB on Netezza 7 years of historical data % annual data growth “NYSE … has replaced an Oracle IO relational database with a data warehousing appliance from Netezza, allowing it to conduct rapid searches of 650 terabytes of data.” ComputerWeekly.com Source: 29

30 Predicts what shoppers are likely to buy in future visits
Smart Predicts what shoppers are likely to buy in future visits Coupon redemption rates as high as 25% “Because of (Netezza’s) in-database technology, we believe we'll be able to do 600 predictive models per year (10X as many as before) with the same staff." Eric Williams, CIO and executive VP 30

31 Todos prometem, mas... nós provamos!
Nós provamos que somos simples Nós provamos que entregamos performance Nós provamos dentro do seu ambiente Nós provamos que nos integramos com suas ferramentas Nós provamos que somos “fáceis de fazer negócio” Nós provamos que temos o menor TCO Nós provamos Business Value

32 Listar os passos de uma PoC
1- Definir com cliente, os testes a serem realizados 2- Obter as queries e as DDLs a serem usadas na PoC 3- Criar as tabelas 4- Testes de carga, leitura, atualização e concorrência 5- Comparar as consultas no sistema atual e no Netezza 6- Duração de 1 semana (2 semanas no máximo) 32

33 Indice de sucesso nas PoCs:
86% This is a number we like to boast about A number that we hope you’ll come to cherish as well and help us maintain and grow in the future This is our win-rate against Oracle, both historic and current, as of last quarter … with and without Exadata! In fact, even when we lost deals, we lost them on business grounds … against Oracle ELAs and business relationships .. and not on the technical merits of their products Obviously our obsession has paid off very well Click to proceed The acquisition is naturally making Larry nervous He knows that the success of Exadata is key to his ambitions against IBM He also knows that if he couldn’t beat Netezza as a standalone company, he doesn’t stand a chance with the combination of Netezza and IBM When it comes to data warehousing, we have the right technology leadership, experience, proven customer successes and the right formula for winning … every single time! One of “The five most important M&A Deals of 2010” - Wall Street Journal 33

34 Retail / Consumer Products
Digital Media Financial Services Governo Health & Life Sciences Retail / Consumer Products A Company is judged by the Company they keep. Those were just a few examples from over 500 Netezza customers Our customers span a variety of vertical industries and sizes Telecom Other Page 34 34

35 Obrigado! (slides backup)

36 Oracle Exadata 3636 Oracle Exadata Results In Netezza TwinFin
Netezza’s Competitive Advantage Architecture Two layer: Clustered SMP DB Layer (RAC) Shared disk MPP Storage Layer Compromised Performance True MPP with FPGA acceleration of processing in each MPP node Best architecture for DW and advanced analytics due to minimization of contention/bottlenecks Speed Tuned for OLTP (e.g. FlashCache) RAC unfit for DW workloads Poor DW Performance Appliance tuned for DW and advanced analytics Highest DW performance Operational Simplicity Simplicity Complexity of Oracle Real Application Clusters (RAC) Constant tuning for performance Complex Administration True Appliance with HW/SW created to provide high performance for DW No tuning More time spent delivering business value rather than tuning for acceptable performance Smart Very limited push-down of analytics RAC bottleneck for analytic performance Poor Analytic Performance Push down of many diverse analytics (SAS, R, Gnu, etc.) through iClass Ability to accelerate the analytics used by many prospects Costs Acquisition cost can exceed $7M per rack Hardware $1M Software is more than $6M! High maintenance and software subscription Continuing high admin costs High Total Cost of Ownership Low, transparent initial cost Simple install requires no additional professional services Standard maintenance includes hw /sw support and sw upgrades Easily understood, predictable costs Minimal “extra” services so easier to budget for Netezza 3636

37 Analysis Summary: Oracle Exadata Database Machine
Exadata is Limited in the Processing It Does. Won’t Handle: Complex joins Distinct aggregation Analytical functions Most Work Still Done on Oracle Database Server Lots of movement of data Loss of Performance Oracle Says Exadata Can Do OLTP or DW or Both At the Same Time Vastly different workloads requiring vastly different tuning Netezza customers report that Exadata poor at DW and analytic 37

38 Query Throughput ≠ Scan Rate
Oracle Exadata throws together the very fast hardware and hopes it produces fast results. Exadata offers very fast scan rates but that just means it can get data off the disks quickly. Overall query throughput also relies on the speed of all the other components, including the software Oracle Exadata can be very fast for simple queries but gets slower with increasing complexity Netezza is designed for balance – it works fast for all query types 3838

39 Netezza’s Advantages over Oracle
Oracle RAC is still Oracle RAC. It is still: Complex – needs to be tuned Temperamental – needs retuning for different configurations Difficult – needs specialized skills and constant maintenance Netezza is much easier. With hardware and software optimized for data warehouse applications, there is: No need for labor-intensive tuning No requirements for partitioning, indexing or building cubes Database Machine is a Resource Hog For a full rack Oracle Exadata Database Machine, you will need to supply at least 90 IP addresses (22 IP addresses for the InfiniBand network, 68 IP addresses for Ethernet, assuming a single cluster), and a minimum of 10 network drops (with 50+ reported as being typical). In contrast, a Netezza TwinFin-12 requires 5 IP addresses and 4 network drops. The core Netezza theme of simplicity is reflected in installation as in operation. 39

40 TwinFin™ 24 Specification
16 (8*2) Disk Enclosures 192 (96*2) 1TB SAS Drives (8 hot spares) RAID 1 Mirroring 2 Hosts (Active-Passive): 24 Cores (Quad-Core Intel 2.6 GHz) 96 GB Memory 4x146 GB SAS Drives Red Hat Linux 5 64-bit 10G Internal Network 24 Netezza S-Blades: 192 Core’s ( Intel Quad-Core 2.5 GHz) 192 FPGA’s ( 125 MHz ) 384 GB DDR2 RAM (1+TB compressed) Linux 64-bit Kernel User Data Capacity: 250 TB Data Scan Speed: 290 TB/hr Load Speed (per system): 2.0 TB/hr Power/Rack: ,400 Watts Cooling/Rack: ,500 BTU/Hour

41 Compress Engine in Action
On Data Load Rows separated into columnar streams Each stream independently compiled Field instructions applied to block headers Compressed data maintains row-based structure On Data Scan/Query FPGA executes field instructions to decompile at wire speed Data re-assembled into rows for other FAST Engines processing 41

42 Workload Management Controls: Guaranteed Resource Allocation
42

43 Default Workload Management: Short Query Bias
Short Query Bias (SQB) Short queries prioritized ahead of longer running queries Real-time responses to users performing short queries Invaluable feature for large mixed-workload environments 8 Items or Less Full Carts Here Full Carts Here 43

44 GRA Test: Fidelity to User Settings
44