O slideshow foi denunciado.
Utilizamos seu perfil e dados de atividades no LinkedIn para personalizar e exibir anúncios mais relevantes. Altere suas preferências de anúncios quando desejar.
IBM Netezza TwinFin ® Líder em Appliances para Data Warehouse  Silvio Ferrari IBM Netezza Systems Engineer [email_address]
Conteúdo Integrate &  Cleanses Dados Estruturados Analisar Integrar Governança Dados Aplicações Transacionais & Colaborati...
Verdadeiros Appliances <ul><li>Dispositivos especializados </li></ul><ul><li>Otimizados para um propósito </li></ul><ul><l...
A Simplicidade de um Appliance Netezza
Carregando dados no Appliance IBM Netezza Integração de dados Inserindo <ul><ul><li>Ab Initio </li></ul></ul><ul><ul><li>B...
Consultando o Appliance IBM Netezza Reporting e Análise <ul><ul><li>Actuate </li></ul></ul><ul><ul><li>Business Objects/SA...
A arquitetura IBM Netezza AMPP™  ( parte de Hardware ) Analíticos Avançados Loader ETL BI Applicações FPGA Memory CPU FPGA...
Servidores Blade CPUs Memória
Acelerador IBM Netezza Database CPUs Memória FPGA
Nosso segredo: FPGA CPU Descomprime Elimina colunas não usadas Restringe Visibilidade Operações complexas: ∑ Joins, Aggs, ...
O S-Blade™ IBM Netezza
Arquitetura IBM Netezza TwinFin™ Hardware+Software Otimizados Projetado (e não simplesmente adaptado) para tarefas analíti...
Simplicidade do Appliance IBM Netezza  ( Software ) <ul><li>dbspace/tablespace: não há sizing ou configuração </li></ul><u...
Complexidade  versus  Simplicidade IBM Netezza    Criando um database: 0.  CREATE DATABASE TEST LOGFILE 'E:OraDataTESTLOG1...
<ul><li>ORACLE </li></ul><ul><li>CREATE TABLE &quot;MRDWDDM&quot;.&quot;RDWF_DDM_ROOMS_SOLD&quot; (&quot;ID_PROPERTY&quot;...
Complexidade Tradicional versus a Simplicidade Netezza (RDBMS 101) CREATE TABLE EDW_PROD.EDW_RESPD_EXPSR_MIN_FACT ( RPT_PE...
Comparação de requerimentos de redes  (internas e externas) Total:  9  endereços IP Total:  90  endereços IP 4  network dr...
Monitorando a distribuição dos dados com NzAdmin <ul><li>Uma má distribuição. </li></ul><ul><li>O usuário escolheu a(s) co...
Uma boa Distribuição: 2.2 Trilhões de Registros
Monitoração: Distribuição homogênea dos dados no sistema <ul><li>Análise de SKEW com relação ao sistema </li></ul>Deve hav...
Backup e Restore <ul><li>Integração e certificação com ferramentas líderes de mercado:  </li></ul><ul><ul><li>Simplifica i...
The IBM Netezza TwinFin™ - Expansão Em caso de expansão: - um novo sistema completo é enviado - dados migrados  ONLINE - I...
i-Class: Analytics Without Constraints <ul><li>Analyze wider and deeper data </li></ul><ul><ul><li>Additional dimensions <...
Advanced Analytics with TwinFin i-Class SAS, SPSS R, S+ SQL SQL Fraud Detection Demand Forecasting
Simples de Instalar e Operar <ul><li>Operações </li></ul><ul><ul><li>Simplesmente carregue e use… é um appliance! </li></u...
Família de Appliances para todo o ciclo de gerenciamento: Skimmer Sistemas de Desenvolvimento e Testes 1 TB to 10 TB TwinF...
15,000 users running 800,000+ queries per day 50X faster than before Speed Source: http:// www.youtube.com/watch?v =yOwnX1...
Simplicity  200X faster than Oracle system ROI in less than 3 months Up and running 6 months before having any training DA...
Scalability Source:   http://www.computerweekly.com/Articles/2008/04/14/230265/NYSE-improves-data-management-with-dataware...
Smart Coupon redemption rates as high as 25% Predicts what shoppers are likely to buy in future visits “ Because of (Netez...
Todos prometem, mas... nós provamos! <ul><li>Nós provamos que somos  simples </li></ul><ul><li>Nós provamos que entregamos...
Listar os passos de uma PoC <ul><li>1- Definir com cliente, os testes a serem realizados </li></ul><ul><li>2- Obter as que...
Indice de sucesso nas PoCs: 86% One of “ The five most important M&A Deals of 2010 ” -  Wall Street Journal
Page  Digital Media Financial Services Governo Health & Life Sciences Retail / Consumer Products Telecom Other
Obrigado! (slides backup)
Oracle Exadata Oracle Exadata Results In Netezza TwinFin Netezza’s Competitive Advantage Architecture <ul><li>Two layer: <...
Analysis Summary:   Oracle Exadata Database Machine <ul><li>Exadata is Limited in the Processing It Does. Won’t Handle:   ...
Query Throughput  ≠  Scan Rate <ul><li>Oracle Exadata throws together the very fast hardware and hopes it produces fast re...
Netezza’s Advantages over Oracle <ul><li>Oracle RAC is still Oracle RAC. It is still:  </li></ul><ul><ul><li>Complex – nee...
TwinFin™ 24 Specification <ul><li>16 (8*2) Disk Enclosures </li></ul><ul><li>192 (96*2) 1TB SAS Drives  </li></ul><ul><li>...
Compress Engine in Action <ul><li>On Data Load </li></ul><ul><li>Rows separated into columnar streams </li></ul><ul><li>Ea...
Workload Management Controls: Guaranteed Resource Allocation
Default Workload Management: Short Query Bias <ul><li>Short Query Bias (SQB) </li></ul><ul><ul><li>Short queries prioritiz...
GRA Test: Fidelity to User Settings
Próximos SlideShares
Carregando em…5
×

Netezza technicaloverviewportugues

1.676 visualizações

Publicada em

Publicada em: Tecnologia, Negócios

Netezza technicaloverviewportugues

  1. 1. IBM Netezza TwinFin ® Líder em Appliances para Data Warehouse Silvio Ferrari IBM Netezza Systems Engineer [email_address]
  2. 2. Conteúdo Integrate & Cleanses Dados Estruturados Analisar Integrar Governança Dados Aplicações Transacionais & Colaborativas Gerenciar Informação Streaming Aplicações Analíticas de Negócio Streams Big Data Data Warehouses Fontes de informação Externas www Qualidade Gerenciamento de Lifecycle Segurança & Privacidade Netezza, IM e BAO Data Warehouse Appliances Master Data
  3. 3. Verdadeiros Appliances <ul><li>Dispositivos especializados </li></ul><ul><li>Otimizados para um propósito </li></ul><ul><li>Solução completa </li></ul><ul><li>Instalação rápida </li></ul><ul><li>Operação muito simples </li></ul><ul><li>Interfaces padrão de mercado </li></ul><ul><li>Baixo custo </li></ul><ul><li>Netezza anuncia servidor em 2002 </li></ul><ul><li>Está no melhor quadrante do Gartner desde 2008 </li></ul><ul><li>2008 Data Warehouse Database Management Systems Magic Quadrant report released on December 23, 2008 </li></ul>
  4. 4. A Simplicidade de um Appliance Netezza
  5. 5. Carregando dados no Appliance IBM Netezza Integração de dados Inserindo <ul><ul><li>Ab Initio </li></ul></ul><ul><ul><li>Business Objects/SAP </li></ul></ul><ul><ul><li>Composite Software </li></ul></ul><ul><ul><li>Expressor Software </li></ul></ul><ul><ul><li>GoldenGate Software (Oracle) </li></ul></ul><ul><ul><li>Informatica </li></ul></ul><ul><ul><li>IBM Information Server </li></ul></ul><ul><ul><li>Sunopsis (Oracle) </li></ul></ul><ul><ul><li>WisdomForce </li></ul></ul><ul><ul><li>... e outras mais.... </li></ul></ul>SQL ODBC JDBC OLE-DB
  6. 6. Consultando o Appliance IBM Netezza Reporting e Análise <ul><ul><li>Actuate </li></ul></ul><ul><ul><li>Business Objects/SAP </li></ul></ul><ul><ul><li>Cognos (IBM) </li></ul></ul><ul><ul><li>Information Builders </li></ul></ul><ul><ul><li>Kalido </li></ul></ul><ul><ul><li>KXEN </li></ul></ul><ul><ul><li>MicroStrategy </li></ul></ul><ul><ul><li>Oracle OBIEE </li></ul></ul><ul><ul><li>QlikTech </li></ul></ul><ul><ul><li>Quest Software </li></ul></ul><ul><ul><li>SAS </li></ul></ul><ul><ul><li>SPSS (IBM) </li></ul></ul><ul><ul><li>Unica (IBM) </li></ul></ul><ul><ul><li>... e outras mais.... </li></ul></ul>extraindo SQL ODBC JDBC OLE-DB
  7. 7. A arquitetura IBM Netezza AMPP™ ( parte de Hardware ) Analíticos Avançados Loader ETL BI Applicações FPGA Memory CPU FPGA Memory CPU FPGA Memory CPU Discos S-Blades™ Rede Interna Netezza Appliance Hosts Host
  8. 8. Servidores Blade CPUs Memória
  9. 9. Acelerador IBM Netezza Database CPUs Memória FPGA
  10. 10. Nosso segredo: FPGA CPU Descomprime Elimina colunas não usadas Restringe Visibilidade Operações complexas: ∑ Joins, Aggs, etc. select DISTRICT, PRODUCTGRP, sum(NRX) from MTHLY_RX_TERR_DATA where MONTH = '20091201' and MARKET = 509123 and SPECIALTY = 'GASTRO' Parte da tabela MTHLY_RX_TERR_DATA (comprimida) where MONTH = '20091201' and MARKET = 509123 and SPECIALTY = 'GASTRO' sum(NRX) select DISTRICT, PRODUCTGRP, sum(NRX)
  11. 11. O S-Blade™ IBM Netezza
  12. 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 Verdadeiro MPP Todos os processadores totalmente utilizados para máxima eficiência e velocidade Analíticos avançados Analíticos complexos executados in-database
  13. 13. Simplicidade do Appliance IBM Netezza ( Software ) <ul><li>dbspace/tablespace: não há sizing ou configuração </li></ul><ul><li>redo/physical/Logical log: não há sizing ou configuração </li></ul><ul><li>page/block de tabelas: não há sizing ou configuração </li></ul><ul><li>extent para tabelas não há sizing ou configuração </li></ul><ul><li>Temp Space: não há alocação ou monitoração </li></ul><ul><li>dbspaces: não há decisões para nível RAID </li></ul><ul><li>Logical Volume: não há criação de files </li></ul><ul><li>OS kernel: não há alterações </li></ul><ul><li>OS kernel: não há níveis de patch requeridos </li></ul><ul><li>Sessões JAD para configurar host/network/storage não requeridas </li></ul>Administração de storage desnecessária Sem índices ou ajustes Sem instalação de software Passos da instalação: - conectar energia elétrica - rodar testes (8h) - entregar servidor ao cliente DBAs se tornam Gerenciadores de Dados, em vez de administradores de banco de dados
  14. 14. Complexidade versus Simplicidade IBM Netezza Criando um database: 0. CREATE DATABASE TEST LOGFILE 'E:OraDataTESTLOG1TEST.ORA' SIZE 2M, 'E:OraDataTESTLOG2TEST.ORA' SIZE 2M, 'E:OraDataTESTLOG3TEST.ORA' SIZE 2M, 'E:OraDataTESTLOG4TEST.ORA' SIZE 2M, 'E:OraDataTESTLOG5TEST.ORA' SIZE 2M EXTENT MANAGEMENT LOCAL MAXDATAFILES 100 DATAFILE 'E:OraDataTESTSYS1TEST.ORA' SIZE 50 M DEFAULT TEMPORARY TABLESPACE temp TEMPFILE 'E:OraDataTESTTEMP.ORA' SIZE 50 M UNDO TABLESPACE undo DATAFILE 'E:OraDataTESTUNDO.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;
  15. 15. <ul><li>ORACLE </li></ul><ul><li>CREATE TABLE &quot;MRDWDDM&quot;.&quot;RDWF_DDM_ROOMS_SOLD&quot; (&quot;ID_PROPERTY&quot; NUMBER(5, </li></ul><ul><li>0) NOT NULL ENABLE, &quot;ID_DATE_STAY&quot; NUMBER(5, 0) NOT NULL ENABLE, </li></ul><ul><li>&quot;CD_ROOM_POOL&quot; CHAR(4) NOT NULL ENABLE, &quot;CD_RATE_PGM&quot; CHAR(4) NOT </li></ul><ul><li>NULL ENABLE, &quot;CD_RATE_TYPE&quot; CHAR(1) NOT NULL ENABLE, </li></ul><ul><li>&quot;CD_MARKET_SEGMENT&quot; CHAR(2) NOT NULL ENABLE, &quot;ID_CONFO_NUM_ORIG&quot; </li></ul><ul><li>NUMBER(9, 0) NOT NULL ENABLE, &quot;ID_CONFO_NUM_CUR&quot; NUMBER(9, 0) NOT </li></ul><ul><li>NULL ENABLE, &quot;ID_DATE_CREATE&quot; NUMBER(5, 0) NOT NULL ENABLE, </li></ul><ul><li>&quot;ID_DATE_ARRIVAL&quot; NUMBER(5, 0) NOT NULL ENABLE, &quot;ID_DATE_DEPART&quot; </li></ul><ul><li>NUMBER(5, 0) NOT NULL ENABLE, &quot;QY_ROOMS&quot; NUMBER(5, 0) NOT NULL </li></ul><ul><li>ENABLE, &quot;CU_REV_PROJ_NET_LOCAL&quot; NUMBER(21, 3) NOT NULL ENABLE, </li></ul><ul><li>&quot;CU_REV_PROJ_NET_USD&quot; NUMBER(21, 3) NOT NULL ENABLE, </li></ul><ul><li>&quot;QY_DAYS_STAY_CUR&quot; NUMBER(3, 0) NOT NULL ENABLE, &quot;CD_BOOK_SOURCE&quot; </li></ul><ul><li>CHAR(1) NOT NULL ENABLE) PCTFREE 5 PCTUSED 95 INITRANS 4 MAXTRANS 255 </li></ul><ul><li>STORAGE( FREELISTS 6) TABLESPACE &quot;DDM_ROOMS_SOLD_DATA&quot; NOLOGGING </li></ul><ul><li>PARTITION BY RANGE (&quot;ID_PROPERTY&quot; ) (PARTITION &quot;PART1&quot; VALUES LESS </li></ul><ul><li>THAN (600) PCTFREE 5 PCTUSED 95 INITRANS 4 MAXTRANS 255 </li></ul><ul><li>STORAGE(INITIAL 16777216 FREELISTS 6 FREELIST GROUPS 1) TABLESPACE </li></ul><ul><li>&quot;DDM_ROOMS_SOLD_DATA&quot; NOLOGGING NOCOMPRESS, PARTITION &quot;PART2&quot; VALUES </li></ul><ul><li>LESS THAN (1200) PCTFREE 5 PCTUSED 95 INITRANS 4 MAXTRANS 255 </li></ul><ul><li>STORAGE(INITIAL 16777216 FREELISTS 6 FREELIST GROUPS 1) TABLESPACE </li></ul><ul><li>&quot;DDM_ROOMS_SOLD_DATA&quot; NOLOGGING NOCOMPRESS, PARTITION &quot;PART3&quot; VALUES </li></ul><ul><li>LESS THAN (1800) PCTFREE 5 PCTUSED 95 INITRANS 4 MAXTRANS 255 </li></ul><ul><li>STORAGE(INITIAL 16777216 FREELISTS 6 FREELIST GROUPS 1) TABLESPACE </li></ul><ul><li>&quot;DDM_ROOMS_SOLD_DATA&quot; NOLOGGING NOCOMPRESS, PARTITION &quot;PART4&quot; VALUES </li></ul><ul><li>LESS THAN (2400) PCTFREE 5 PCTUSED 95 INITRANS 4 MAXTRANS 255 </li></ul><ul><li>STORAGE(INITIAL 16777216 FREELISTS 6 FREELIST GROUPS 1) TABLESPACE </li></ul><ul><li>&quot;DDM_ROOMS_SOLD_DATA&quot; NOLOGGING NOCOMPRESS, PARTITION &quot;PART5&quot; VALUES </li></ul><ul><li>LESS THAN (3000) PCTFREE 5 PCTUSED 95 INITRANS 4 MAXTRANS 255 </li></ul><ul><li>STORAGE(INITIAL 16777216 FREELISTS 6 FREELIST GROUPS 1) TABLESPACE </li></ul><ul><li>&quot;DDM_ROOMS_SOLD_DATA&quot; NOLOGGING NOCOMPRESS, PARTITION &quot;PART6&quot; VALUES </li></ul><ul><li>LESS THAN (MAXVALUE) PCTFREE 5 PCTUSED 95 INITRANS 4 MAXTRANS 255 </li></ul><ul><li>STORAGE(INITIAL 16777216 FREELISTS 6 FREELIST GROUPS 1) TABLESPACE </li></ul><ul><li>&quot;DDM_ROOMS_SOLD_DATA&quot; NOLOGGING NOCOMPRESS ) ; </li></ul>Simplicidade Netezza: criando uma tabela ORACLE Indexes CREATE INDEX &quot;MRDWDDM&quot;.&quot;RDWF_DDM_ROOMS_SOLD_IDX1&quot; ON &quot;RDWF_DDM_ROOMS_SOLD&quot; (&quot;ID_PROPERTY&quot; , &quot;ID_DATE_STAY&quot; , &quot;CD_ROOM_POOL&quot; , &quot;CD_RATE_PGM&quot; , &quot;CD_RATE_TYPE&quot; , &quot;CD_MARKET_SEGMENT&quot; ) PCTFREE 10 INITRANS 6 MAXTRANS 255 STORAGE( FREELISTS 10) TABLESPACE &quot;DDM_DATAMART_INDEX_L&quot; NOLOGGING PARALLEL ( DEGREE 4 INSTANCES 1) LOCAL(PARTITION &quot;PART1&quot; PCTFREE 10 INITRANS 6 MAXTRANS 255 STORAGE(INITIAL 4194304 NEXT 4259840 MINEXTENTS 1 MAXEXTENTS 100000 PCTINCREASE 0 FREELISTS 10 FREELIST GROUPS 1 BUFFER_POOL DEFAULT) TABLESPACE &quot;DDM_DATAMART_INDEX_L&quot; NOLOGGING, PARTITION &quot;PART2&quot; PCTFREE 10 INITRANS 6 MAXTRANS 255 STORAGE(INITIAL 4194304 NEXT 4259840 MINEXTENTS 1 MAXEXTENTS 100000 PCTINCREASE 0 FREELISTS 10 FREELIST GROUPS 1 BUFFER_POOL DEFAULT) TABLESPACE &quot;DDM_DATAMART_INDEX_L&quot; NOLOGGING, PARTITION &quot;PART3&quot; PCTFREE 10 INITRANS 6 MAXTRANS 255 STORAGE(INITIAL 4194304 NEXT 4259840 MINEXTENTS 1 MAXEXTENTS 100000 PCTINCREASE 0 FREELISTS 10 FREELIST GROUPS 1 BUFFER_POOL DEFAULT) TABLESPACE &quot;DDM_DATAMART_INDEX_L&quot; NOLOGGING, PARTITION &quot;PART4&quot; PCTFREE 10 INITRANS 6 MAXTRANS 255 STORAGE(INITIAL 4194304 NEXT 4259840 MINEXTENTS 1 MAXEXTENTS 100000 PCTINCREASE 0 FREELISTS 10 FREELIST GROUPS 1 BUFFER_POOL DEFAULT) TABLESPACE &quot;DDM_DATAMART_INDEX_L&quot; NOLOGGING, PARTITION &quot;PART5&quot; PCTFREE 10 INITRANS 6 MAXTRANS 255 STORAGE(INITIAL 4194304 NEXT 4259840 MINEXTENTS 1 MAXEXTENTS 100000 PCTINCREASE 0 FREELISTS 10 FREELIST GROUPS 1 BUFFER_POOL DEFAULT) TABLESPACE &quot;DDM_DATAMART_INDEX_L&quot; NOLOGGING, PARTITION &quot;PART6&quot; PCTFREE 10 INITRANS 6 MAXTRANS 255 STORAGE(INITIAL 4194304 NEXT 4259840 MINEXTENTS 1 MAXEXTENTS 100000 PCTINCREASE 0 FREELISTS 10 FREELIST GROUPS 1 BUFFER_POOL DEFAULT) TABLESPACE &quot;DDM_DATAMART_INDEX_L&quot; NOLOGGING ) ; ORACLE Bitmap index CREATE BITMAP INDEX &quot;CRDBO&quot;.&quot;SNAPSHOT_MONTH_IDX13&quot; ON &quot;SNAPSHOT_OPPTY_MONTH_HIST&quot; (&quot;SNAPSHOT_YEAR&quot; ) PCTFREE 10 INITRANS 2 MAXTRANS 255 STORAGE(INITIAL 4194304 NEXT 4194304 MINEXTENTS 2 MAXEXTENTS 2147483645 PCTINCREASE 0 FREELISTS 1 FREELIST GROUPS 1 BUFFER_POOL DEFAULT) TABLESPACE &quot;SFA_DATAMART_INDEX&quot; NOLOGGING ; ORACLE Table Clusters CREATE CLUSTER &quot;MRDW&quot;.&quot;CT_INTRMDRY_CAL&quot; (&quot;ID_YEAR_CAL&quot; NUMBER(4, 0), &quot;ID_MONTH_CAL&quot; NUMBER(2, 0), &quot;ID_PROPERTY&quot; NUMBER(5, 0)) SIZE 16384 PCTFREE 10 PCTUSED 90 INITRANS 3 MAXTRANS 255 STORAGE(INITIAL 83886080 NEXT 41943040 MINEXTENTS 1 MAXEXTENTS 1017 PCTINCREASE 0 FREELISTS 4 FREELIST GROUPS 1 BUFFER_POOL RECYCLE) TABLESPACE &quot;TSS_FACT&quot; ; 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; <ul><li>Sem indexes </li></ul><ul><li>Sem Admininstração ou ajustes </li></ul><ul><li>Distribua os dados aleatoriamente, ou por Colunas </li></ul>
  16. 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) ) 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; 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 TABLESPACE AT_EDW_REXMIN PCTFREE 10 INITRANS 1 MAXTRANS 255 STORAGE ( INITIAL 96K NEXT 96K MINEXTENTS 1 MAXEXTENTS UNLIMITED PCTINCREASE 0 BUFFER_POOL DEFAULT ), PARTITION RP0001 VALUES LESS THAN (2) NOLOGGING NOCOMPRESS TABLESPACE AT_EDW_REXMIN PCTFREE 10 INITRANS 1 MAXTRANS 255 STORAGE ( INITIAL 96K NEXT 96K MINEXTENTS 1 MAXEXTENTS UNLIMITED PCTINCREASE 0 BUFFER_POOL DEFAULT ), PARTITION RP0002 VALUES LESS THAN (3) NOLOGGING NOCOMPRESS TABLESPACE AT_EDW_REXMIN PCTFREE 10 INITRANS 1 MAXTRANS 255 STORAGE ( INITIAL 96K NEXT 96K MINEXTENTS 1 MAXEXTENTS UNLIMITED PCTINCREASE 0 BUFFER_POOL DEFAULT ), … … PLUS DDL FOR 513 MORE PARTITIONS 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 TABLESPACE AI_EDW_REXMIN PCTFREE 10 INITRANS 2 MAXTRANS 255 STORAGE ( INITIAL 96K NEXT 96K MINEXTENTS 1 MAXEXTENTS UNLIMITED PCTINCREASE 0 BUFFER_POOL DEFAULT ), PARTITION RP0001 NOLOGGING NOCOMPRESS TABLESPACE AI_EDW_REXMIN PCTFREE 10 INITRANS 2 MAXTRANS 255 STORAGE ( INITIAL 96K NEXT 96K MINEXTENTS 1 MAXEXTENTS UNLIMITED PCTINCREASE 0 BUFFER_POOL DEFAULT ), PARTITION RP0002 NOLOGGING TABLESPACE AI_EDW_REXMIN PCTFREE 10 INITRANS 2 MAXTRANS 255 STORAGE ( INITIAL 96K NEXT 96K MINEXTENTS 1 MAXEXTENTS UNLIMITED PCTINCREASE 0 BUFFER_POOL DEFAULT ), … … PLUS DDL FOR 512 MORE PARTITIONS 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 TABLESPACE AI_EDW_REXMIN PCTFREE 10 INITRANS 2 MAXTRANS 255 STORAGE ( INITIAL 96K NEXT 96K MINEXTENTS 1 MAXEXTENTS UNLIMITED PCTINCREASE 0 BUFFER_POOL DEFAULT ), PARTITION RP0001 NOLOGGING TABLESPACE AI_EDW_REXMIN PCTFREE 10 INITRANS 2 MAXTRANS 255 STORAGE ( INITIAL 96K NEXT 96K MINEXTENTS 1 MAXEXTENTS UNLIMITED PCTINCREASE 0 BUFFER_POOL DEFAULT ), … … 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 TABLESPACE AI_EDW_REXMIN PCTFREE 10 INITRANS 2 MAXTRANS 255 STORAGE ( INITIAL 96K NEXT 96K MINEXTENTS 1 MAXEXTENTS UNLIMITED PCTINCREASE 0 BUFFER_POOL DEFAULT ), PARTITION RP0001 NOLOGGING TABLESPACE AI_EDW_REXMIN PCTFREE 10 INITRANS 2 MAXTRANS 255 STORAGE ( INITIAL 96K NEXT 96K MINEXTENTS 1 MAXEXTENTS UNLIMITED PCTINCREASE 0 BUFFER_POOL DEFAULT ), … … 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 TABLESPACE AI_EDW_REXMIN PCTFREE 10 INITRANS 2 MAXTRANS 255 STORAGE ( INITIAL 96K NEXT 96K MINEXTENTS 1 MAXEXTENTS UNLIMITED PCTINCREASE 0 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 TABLESPACE AI_EDW_REXMIN PCTFREE 10 INITRANS 2 MAXTRANS 255 STORAGE ( INITIAL 96K NEXT 96K MINEXTENTS 1 MAXEXTENTS UNLIMITED PCTINCREASE 0 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 + 515 PARTITIONS Oracle: 34,500 KB de DDLs Netezza: 250 KB de DDLs
  17. 17. Comparação de requerimentos de redes (internas e externas) Total: 9 endereços IP Total: 90 endereços IP 4 network drops 10 network drops minimum (with 50+ reported as being typical 5 IP addresses 68 IP addresses for Ethernet (for a single cluster) - 22 IP addresses for the InfiniBand network TwinFin12 (full rack) Exadata (full rack)
  18. 18. Monitorando a distribuição dos dados com NzAdmin <ul><li>Uma má distribuição. </li></ul><ul><li>O usuário escolheu a(s) coluna(s) errada(s) para a distribuição dos dados. </li></ul><ul><li>Nota: Neste caso, o usuário escolheu a primeira coluna da tabela como a coluna de distrubuição. Uma decisão incorreta. </li></ul>
  19. 19. Uma boa Distribuição: 2.2 Trilhões de Registros
  20. 20. Monitoração: Distribuição homogênea dos dados no sistema <ul><li>Análise de SKEW com relação ao sistema </li></ul>Deve haver uma carga de utilização equivalente entre as SPUs
  21. 21. Backup e Restore <ul><li>Integração e certificação com ferramentas líderes de mercado: </li></ul><ul><ul><li>Simplifica integração com as principais ferramentas de backup e restore </li></ul></ul><ul><ul><li>Suporte a X/Open Backup Services API (XBSA) </li></ul></ul><ul><ul><li>Certificação IBM Tivoli Storage Manager (TSM) </li></ul></ul><ul><ul><li>Certificação Veritas NetBackup™ da Symantec </li></ul></ul><ul><li>Backup and Restore Incremental </li></ul><ul><ul><li>Diminui significativamente os tempos de backup comparados ao backup Full </li></ul></ul><ul><ul><li>Disponível no utilitário NZBACKUP </li></ul></ul><ul><ul><li>Restores tipo Full ou parcial </li></ul></ul>Dom Seg Ter Qua Qui Sex Sab Full Dif Dif Cumulativo Dif Dif Dif
  22. 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
  23. 23. i-Class: Analytics Without Constraints <ul><li>Analyze wider and deeper data </li></ul><ul><ul><li>Additional dimensions </li></ul></ul><ul><ul><li>Richer history </li></ul></ul>Big Data Big Math <ul><li>Increase computational intensity </li></ul><ul><ul><li>More complex models </li></ul></ul><ul><ul><li>Faster execution for results </li></ul></ul>
  24. 24. Advanced Analytics with TwinFin i-Class SAS, SPSS R, S+ SQL SQL Fraud Detection Demand Forecasting
  25. 25. Simples de Instalar e Operar <ul><li>Operações </li></ul><ul><ul><li>Simplesmente carregue e use… é um appliance! </li></ul></ul><ul><ul><li>Instalação em ~2 dias! </li></ul></ul><ul><ul><li>Fácil de avaliar e funciona como anunciado! </li></ul></ul><ul><li>Desenvolvedores BI & DBAs – mais ágeis </li></ul><ul><ul><li>Sem configuração ou modelagem física </li></ul></ul><ul><ul><li>Sem índices ou ajustes – performance imediata </li></ul></ul><ul><ul><li>Agnóstico a modelos de dados </li></ul></ul><ul><ul><li>Data Architects / DBA focam nos negócios, não na modelagem física </li></ul></ul><ul><li>Desenvolvedores ETL </li></ul><ul><ul><li>Tabelas de agregação não necessárias – lógica de ETL simplificada </li></ul></ul><ul><ul><li>Cargas e transformações mais rápidas </li></ul></ul><ul><li>Analistas de Negócio </li></ul><ul><ul><li>Análise “Linha de Pensamento”– 10 a 100x mais rápida </li></ul></ul><ul><ul><li>Consultas ad hoc – sem ajustes, sem índices </li></ul></ul><ul><ul><li>Consultas complexas a grandes datasets </li></ul></ul><ul><ul><li>Menor latencia – cargas e consultas simultâneas </li></ul></ul><ul><ul><li>processamento OnStream a centenas de nodes </li></ul></ul>
  26. 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. 27. 15,000 users running 800,000+ queries per day 50X faster than before Speed Source: http:// www.youtube.com/watch?v =yOwnX14nLrE&feature= player_embedded “… 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
  28. 28. Simplicity 200X faster than Oracle system ROI in less than 3 months Up and running 6 months before having any training DAYS WEEKS MONTHS “ Allowing the business users access to the Netezza box was what sold it.” Steve Taff, Executive Dir. of IT Services
  29. 29. Scalability Source: http://www.computerweekly.com/Articles/2008/04/14/230265/NYSE-improves-data-management-with-datawarehousing.htm 1 PB on Netezza 7 years of historical data 100-200% 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
  30. 30. Smart Coupon redemption rates as high as 25% Predicts what shoppers are likely to buy in future visits “ 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.&quot; Eric Williams, CIO and executive VP
  31. 31. Todos prometem, mas... nós provamos! <ul><li>Nós provamos que somos simples </li></ul><ul><li>Nós provamos que entregamos performance </li></ul><ul><li>Nós provamos dentro do seu ambiente </li></ul><ul><li>Nós provamos que nos integramos com suas ferramentas </li></ul><ul><li>Nós provamos que somos “ fáceis de fazer negócio ” </li></ul><ul><li>Nós provamos que temos o menor TCO </li></ul><ul><li>Nós provamos Business Value </li></ul>
  32. 32. Listar os passos de uma PoC <ul><li>1- Definir com cliente, os testes a serem realizados </li></ul><ul><li>2- Obter as queries e as DDLs a serem usadas na PoC </li></ul><ul><li>3- Criar as tabelas </li></ul><ul><li>4- Testes de carga, leitura, atualização e concorrência </li></ul><ul><li>5- Comparar as consultas no sistema atual e no Netezza </li></ul><ul><li>6- Duração de 1 semana (2 semanas no máximo) </li></ul>
  33. 33. Indice de sucesso nas PoCs: 86% One of “ The five most important M&A Deals of 2010 ” - Wall Street Journal
  34. 34. Page Digital Media Financial Services Governo Health & Life Sciences Retail / Consumer Products Telecom Other
  35. 35. Obrigado! (slides backup)
  36. 36. Oracle Exadata Oracle Exadata Results In Netezza TwinFin Netezza’s Competitive Advantage Architecture <ul><li>Two layer: </li></ul><ul><ul><li>Clustered SMP DB Layer (RAC) </li></ul></ul><ul><ul><li>Shared disk MPP Storage Layer </li></ul></ul>Compromised Performance <ul><li>True MPP with FPGA acceleration of processing in each MPP node </li></ul><ul><li>Best architecture for DW and advanced analytics due to minimization of contention/bottlenecks </li></ul>Speed <ul><li>Tuned for OLTP (e.g. FlashCache) </li></ul><ul><li>RAC unfit for DW workloads </li></ul>Poor DW Performance <ul><li>Appliance tuned for DW and advanced analytics </li></ul><ul><li>Highest DW performance </li></ul><ul><li>Operational Simplicity </li></ul>Simplicity <ul><li>Complexity of Oracle Real Application Clusters (RAC) </li></ul><ul><li>Constant tuning for performance </li></ul>Complex Administration <ul><li>True Appliance with HW/SW created to provide high performance for DW </li></ul><ul><li>No tuning </li></ul><ul><li>More time spent delivering business value rather than tuning for acceptable performance </li></ul>Smart <ul><li>Very limited push-down of analytics </li></ul><ul><li>RAC bottleneck for analytic performance </li></ul>Poor Analytic Performance <ul><li>Push down of many diverse analytics (SAS, R, Gnu, etc.) through iClass </li></ul><ul><li>Ability to accelerate the analytics used by many prospects </li></ul>Costs <ul><li>Acquisition cost can exceed $7M per rack </li></ul><ul><ul><li>Hardware $1M </li></ul></ul><ul><ul><li>Software is more than $6M! </li></ul></ul><ul><li>High maintenance and software subscription </li></ul><ul><li>Continuing high admin costs </li></ul>High Total Cost of Ownership <ul><li>Low, transparent initial cost </li></ul><ul><li>Simple install requires no additional professional services </li></ul><ul><li>Standard maintenance includes hw /sw support and sw upgrades </li></ul><ul><li>Easily understood, predictable costs </li></ul><ul><li>Minimal “extra” services so easier to budget for Netezza </li></ul>
  37. 37. Analysis Summary: Oracle Exadata Database Machine <ul><li>Exadata is Limited in the Processing It Does. Won’t Handle: </li></ul><ul><ul><li>Complex joins </li></ul></ul><ul><ul><li>Distinct aggregation </li></ul></ul><ul><ul><li>Analytical functions </li></ul></ul><ul><li>Most Work Still Done on Oracle Database Server </li></ul><ul><ul><li>Lots of movement of data </li></ul></ul><ul><ul><li>Loss of Performance </li></ul></ul><ul><li>Oracle Says Exadata Can Do OLTP or DW or Both At the Same Time </li></ul><ul><ul><li>Vastly different workloads requiring vastly different tuning </li></ul></ul><ul><ul><li>Netezza customers report that Exadata poor at DW and analytic </li></ul></ul>
  38. 38. Query Throughput ≠ Scan Rate <ul><li>Oracle Exadata throws together the very fast hardware and hopes it produces fast results. </li></ul><ul><li>Exadata offers very fast scan rates but that just means it can get data off the disks quickly. </li></ul><ul><li>Overall query throughput also relies on the speed of all the other components, including the software </li></ul><ul><li>Oracle Exadata can be very fast for simple queries but gets slower with increasing complexity </li></ul><ul><li>Netezza is designed for balance – it works fast for all query types </li></ul>
  39. 39. Netezza’s Advantages over Oracle <ul><li>Oracle RAC is still Oracle RAC. It is still: </li></ul><ul><ul><li>Complex – needs to be tuned </li></ul></ul><ul><ul><li>Temperamental – needs retuning for different configurations </li></ul></ul><ul><ul><li>Difficult – needs specialized skills and constant maintenance </li></ul></ul><ul><li>Netezza is much easier. With hardware and software optimized for data warehouse applications, there is: </li></ul><ul><ul><li>No need for labor-intensive tuning </li></ul></ul><ul><ul><li>No requirements for partitioning, indexing or building cubes </li></ul></ul><ul><li>Database Machine is a Resource Hog </li></ul><ul><ul><li>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 ). </li></ul></ul><ul><li>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. </li></ul>
  40. 40. TwinFin™ 24 Specification <ul><li>16 (8*2) Disk Enclosures </li></ul><ul><li>192 (96*2) 1TB SAS Drives </li></ul><ul><li>(8 hot spares) </li></ul><ul><li>RAID 1 Mirroring </li></ul><ul><li>24 Netezza S-Blades: </li></ul><ul><li>192 Core’s ( Intel Quad-Core 2.5 GHz) </li></ul><ul><li>192 FPGA’s ( 125 MHz ) </li></ul><ul><li>384 GB DDR2 RAM (1+TB compressed) </li></ul><ul><li>Linux 64-bit Kernel </li></ul><ul><li>2 Hosts (Active-Passive): </li></ul><ul><li>24 Cores (Quad-Core Intel 2.6 GHz) </li></ul><ul><li>96 GB Memory </li></ul><ul><li>4x146 GB SAS Drives </li></ul><ul><li>Red Hat Linux 5 64-bit </li></ul><ul><li>10G Internal Network </li></ul><ul><li>User Data Capacity: 250 TB </li></ul><ul><li>Data Scan Speed: 290 TB/hr </li></ul><ul><li>Load Speed (per system): 2.0 TB/hr </li></ul><ul><li>Power/Rack: 7,400 Watts </li></ul><ul><li>Cooling/Rack: 25,500 BTU/Hour </li></ul>
  41. 41. Compress Engine in Action <ul><li>On Data Load </li></ul><ul><li>Rows separated into columnar streams </li></ul><ul><li>Each stream independently compiled </li></ul><ul><li>Field instructions applied to block headers </li></ul><ul><li>Compressed data maintains row-based structure </li></ul><ul><li>On Data Scan/Query </li></ul><ul><li>FPGA executes field instructions to decompile at wire speed </li></ul><ul><li>Data re-assembled into rows for other FAST Engines processing </li></ul>
  42. 42. Workload Management Controls: Guaranteed Resource Allocation
  43. 43. Default Workload Management: Short Query Bias <ul><li>Short Query Bias (SQB) </li></ul><ul><ul><li>Short queries prioritized ahead of longer running queries </li></ul></ul><ul><ul><li>Real-time responses to users performing short queries </li></ul></ul><ul><ul><li>Invaluable feature for large mixed-workload environments </li></ul></ul>8 Items or Less Full Carts Here Full Carts Here
  44. 44. GRA Test: Fidelity to User Settings

×