Upload
ibm-srbija
View
69
Download
3
Embed Size (px)
DESCRIPTION
Citation preview
© 2014 IBM Corporation
IBM DB2 Analytics Accelerator for z/OS
DB2 Analytics Accelerator for z/OS, is a highperformance solution designed to work with IBMSystem z and DB2 z/OS to deliver faster analytic queryresponses transparently to users.
Saso Prek, IBM [email protected] Certified IT Specialist,Member of zChampion Team
© 2014 IBM Corporation2
Customer #2 & #3DW queries & some figures
DWH.F_FAKTURA_EE > 531.000.000 rows
DWH.D_MER_MESTO > 1.209.877 rows
SELECT A.JP, E.TAR_SKUP_SIF_NAZ, B.MER_MESTO, C.LETO_MES, SUM(A.KOLICINA)
FROM DWH.F_FAKTURA_EE AS A, DWH.D_MER_MESTO AS B, DWH.D_VRS_PRODUKTA_EE AS D,
DWH.D_CAS AS C INNER JOIN DWH.D_TAR_SKUPINA AS E ON E.ID_TAR_SKUPINA = A.ID_TAR_SKUPINA
WHERE (A.JP = 2) AND (A.ID_CAS = C.ID_CAS) AND (C.LETO IN (2012, 2011, 2010, 2009, 2008))
AND (B.ID_MER_MESTO = A.ID_MER_MESTO)
AND (D.ID_VRS_PRODUKTA_EE = A.ID_VRS_PRODUKTA_EE)
AND (D.GRUPA_NAZ = 'OMREŽNINA')
GROUP BY A.JP, E.TAR_SKUP_SIF_NAZ, B.MER_MESTO, C.LETO_MES
with IDAA << 10 sec
w/o IDAA > 2 hours ( xxx CPU & xxx I/O)
© 2014 IBM Corporation3
IBM DB2 Analytics Accelerator - Product Components
3
10Gb
OSA-Express3
10 GbE
Primary
Backup
CLIENT
Data Studio Foundation
DB2 AnalyticsAccelerator
Admin Plug-in
zEnterprise + zEC12 + zBC12
Data Warehouse applicationDB2 for z/OS enabled for IBM
DB2 Analytics AcceleratorIBM DB2 AnalyticsAcelerator
BladeCenter
NetezzaTechnology
Users/Applications
Network
© 2014 IBM Corporation4
Fast Time to Value
� IBM DB2 Analytics Accelerator (Netezza 1000-12)
� Production ready - 1 person, 2 days
� Table Acceleration Setup … 2 Hours
– DB2 “Add Accelerator”
– Choose a Table for “Acceleration”
– Load the Table (DB2 copy to Netezza)
– Knowledge Transfer
– Query Comparisons
� Initial Load Performance …
�400 GB “Loaded” in 29 Min 570 million rows (Loads of 800GB to 1.3TB/Hr)
� Actual Query Acceleration … 1908x faster
�2 Hours 39 Minutes to 5 Seconds
� CPU Utilization Reduction
�35% to ~0%
Actual customer results, October 2011
© 2014 IBM Corporation55 16 June 2014
Performance & Savings
Accelerating decisions to the speed of business
Queries run faster
• Save CPU resources
• MIPS
• DASD
• People time
• Business opportunities
Actual customer results, October 2011
Times
Faster
Query
Total
Rows
Reviewed
Total
Rows
Returned Hours Sec(s) Hours Sec(s)
Query 1 2,813,571 853,320 2:39 9,540 0.0 5 1,908
Query 2 2,813,571 585,780 2:16 8,220 0.0 5 1,644
Query 3 8,260,214 274 1:16 4,560 0.0 6 760
Query 4 2,813,571 601,197 1:08 4,080 0.0 5 816
Query 5 3,422,765 508 0:57 4,080 0.0 70 58Query 6 4,290,648 165 0:53 3,180 0.0 6 530
Query 7 361,521 58,236 0:51 3,120 0.0 4 780
Query 8 3,425.29 724 0:44 2,640 0.0 2 1,320
Query 9 4,130,107 137 0:42 2,520 0.1 193 13
DB2 Only
DB2 with
IDAA
DB2 Analytics Accelerator: “we had this up and running in days with queries that ran over 1000
times faster”
DB2 Analytics Accelerator: “we expect ROI in less than 4 months”
Advance to 32 minute mark for DB2 Analytics Accelerator section of keynote
© 2014 IBM Corporation66
DB2 Analytics Accelerator
Blending System z and Netezza
technologies to deliver unparalleled,
mixed workload performance for
complex analytic business needs.
More insight from your data
• Fast, predictable response times for
“right-time” analysis
• Complex queries in seconds rather than hours
• Transparent to the application
• Inherits all System z DB2 attributes
• No need to create or maintain indices
• Eliminate query tuning
• Fast deployment and time-to-value
© 2014 IBM Corporation7
IBM DB2 Analytics Accelerator - Product Components
7
10Gb
OSA-Express3
10 GbE
Primary
Backup
CLIENT
Data Studio Foundation
DB2 AnalyticsAccelerator
Admin Plug-in
zEnterprise + zEC12 + zBC12
Data Warehouse applicationDB2 for z/OS enabled for IBM
DB2 Analytics AcceleratorIBM DB2 AnalyticsAcelerator
BladeCenter
NetezzaTechnology
Users/Applications
Network
© 2014 IBM Corporation888
User InterfaceIncremental update UI elements only visible if it has been enabled on the DB2 subsystem via IBM DB2 Analytics Accelerator configuration console
� Start / stop replication process (per subsystem-accelerator pair)
� Enable / disable replication (per table)
� Trace collection
� Information on replication latency and events
© 2014 IBM Corporation9
Deep DB2 Integration within zEnterprise
9
Data
Manager
Buffer
ManagerIRLM
Log
Manager
IBM
DB2
Analytics
Accelerator
Applications DBA Tools, z/OS Console, ...
. . .
Operational Interfaces
(e.g. DB2 Commands)
Application Interfaces
(standard SQL dialects)
z/OS on System z
DB2 for z/OS
Superior availability
reliability, security,
Workload management
Superior
performance on
analytic queries
© 2014 IBM Corporation10
Fast Time to Value
� IBM DB2 Analytics Accelerator (Netezza 1000-12)
� Production ready - 1 person, 2 days
� Table Acceleration Setup … 2 Hours
– DB2 “Add Accelerator”
– Choose a Table for “Acceleration”
– Load the Table (DB2 copy to Netezza)
– Knowledge Transfer
– Query Comparisons
� Initial Load Performance …
�400 GB “Loaded” in 29 Min 570 million rows (Loads of 800GB to 1.3TB/Hr)
� Actual Query Acceleration … 1908x faster
�2 Hours 39 Minutes to 5 Seconds
� CPU Utilization Reduction
�35% to ~0%
Actual customer results, October 2011
© 2014 IBM Corporation11
SMP Hosts
Snippet BladesTM
(S-Blades, SPUs)
Disk Enclosures
IDAA ServerSQL Compiler, Query Plan, OptimizeAdministration2 front/end hosts, IBM 3650M3
clustered active-passive2 Nehalem-EP Quad-core 2.4GHz per host
Processor &streaming DB logicHigh-performance databaseengine streaming joins,aggregations, sorts, etc.e.g. TF12: 12 back/end SPUs
(more details on following charts)
Slice of User DataSwap and Mirror partitionsHigh speed data streamingHigh compression rate EXP3000 JBOD Enclosures
12 x 3.5” 1TB, 7200RPM, SAS (3Gb/s)max 116MB/s (200-500MB/s compressed data)
e.g. TF12:8 enclosures → 96 HDDs
32TB uncompressed user data (→ 128TB)
DB2 Analytics Accelerator V2Powered by Netezza 1000 Appliance
© 2014 IBM Corporation12
EXPLAIN
• DB2 EXPLAIN function is enhanced to provide basic information about accelerator usage
• Whether query qualifies for acceleration and, if not, why
• The access path details associated with the query execution by Netezza are
provided independently of DB2 EXPLAIN by the IDAA Studio.
• For each query (irrespective of the number of query blocks) a row
is inserted in the following tables:
• in both PLAN_TABLE and DSN_QUERYINFO_TABLE, if the query is re-routed
• PLAN_TABLE's ACCESSTYPE column is set to a value of 'A'
• DSN_QUERYINFO_TABLE's QI_DATA column shows the converted query text
• in DSN_QUERYINFO_TABLE only, if the query is not re-routed
• REASON_CODE and QI_DATA columns provide details
• Note that the EXPLAIN tables can be populated with above described information even if there is no accelerator connected to DB2
• Specifying EXPLAINONLY on START ACCEL command does not establish any communications with an actual accelerator, but enables DB2 to consider its presence in the access path selection process
© 2014 IBM Corporation13
Query Execution Process Flow
13
Optimizer
IDA
A D
RD
A R
equesto
r
Application
Application
Interface
Queries executed with IDAA
Queries executed without IDAA
Heartbeat (IDAA availability and performance indicators)
Query execution run-time for
queries that cannot be or should
not be off-loaded to IDAA
SPU
Memory
SPU
Memory
SPU
Memory
SPU
Memory
SM
P H
ost
Heartbeat
DB2 for z/OS
CPU FPGA
CPU FPGA
CPU FPGA
CPU FPGA
CPU FPGA
CPU FPGA
CPU FPGA
CPU FPGA
IDAA
© 2014 IBM Corporation14
Option 1: Full Table Refresh
� Changes in data warehouse tables typically driven by scheduled (nightly or more frequently) ETL process
� Data used for complex reporting based on consistent and validated content (e.g., weekly transaction reporting to the central bank)
� Multiple sources or complex transformations prevent from propagation of incremental changes
� Queries may continue during full table refresh for accelerator
� Full table refresh may be triggered through DB2 stored procedure (scheduled, integrated into ETL process or through GUI)
DB2 z/OS Query OptimizerDB2 z/OS Query Optimizer
Changes / Replacement
DB2 for z/OS databaseDB2 for z/OS database
ET
L P
roce
ss
ET
L P
roce
ss
DB2 native
processing
DB2 native
processingAccelerator
processing
Accelerator
processing
Operational Analytics, Reports, OLAP, …Operational Analytics, Reports, OLAP, …
Continuous
Query
Processing
Continuous
Query
Processing
Full table refresh
© 2014 IBM Corporation15
Option 2: Incremental Update (Controlled Availability)
� Changes in data warehouse tables typically driven by replication or manual updates
– Corrections after a bulk-ETL-load of a data warehouse table
– Continuously changing data (e.g. trickle-feed updates from a transactional system to an ODS)
� Reporting and analysis based on most recent data
� May be combined with Option 1 & 2 (first table refresh and then continue with incremental updates)
� Incremental update can be configured per database table
DB2 z/OS Query OptimizerDB2 z/OS Query Optimizer
Changes
DB2 for z/OS databaseDB2 for z/OS database
DB2 native
processing
DB2 native
processingAccelerator
processing
Accelerator
processing
Operational Analytics, Reports, OLAP, …Operational Analytics, Reports, OLAP, …
Continuous
Query
Processing
Continuous
Query
Processing
Incremental Update
Replic
ation
Replic
ation
Applic
ation
Applic
ation
© 2014 IBM Corporation1616
Connectivity Options
Multiple DB2 systems can connect to a single IDAA
A single DB2 system can connect to multiple IDAAs
• residing in the same LPAR
• residing in different LPARs
• residing in different CECs
• being independent (non-data sharing)
• belonging to the same data sharing group
• belonging to different data sharing groups
Multiple DB2 systems can connect to multiple IDAAs
Full flexibility for DB2 systems:
Better utilization of IDAA resourcesBetter utilization of IDAA resources
ScalabilityScalability
High availabilityHigh availability
DB2 DB2
DB2
DB2 DB2
© 2014 IBM Corporation17
Save Over 95% of Host Disk Space for Historical Data
1Q
2Q
3Q
4Q
1Q
2Q
3Q
4Q
1Q
2Q
3Q
4Q
1Q
2Q
3Q
4Q
1Q
2Q
3Q
4Q
1Q
2Q
3Q
4Q
1Q
2Q
3Q
4Q
Year Year -7Year -2 Year -3 Year -4 Year -5Year -1
Historical Data
Current Data
One Quarter = 3.57% of 7 years of data
One Month = 1.12% of 7 years of data
One month = 2.78% of 3 years of data
© 2014 IBM Corporation1818
Accelerating SAP for more business value
� Enhances the SAP Certified
DB2 for z/OS
� Accelerates SAP NetWeaver BW
� Dramatic decrease in
elapsed time for SAP BW ad-hoc reporting
No Description
Records
Read
Records
Returned
DB2
(sec)
Accelerator
(sec)
Accleration
Factor
1 Simple mass aggregation 17116647 21 117 0.78 150
2 Query #1 + 70% filter 11980812 21 94.2 0.86 110
3 Query #1 + 30% filter 5133708 21 54.8 0.82 67
4 Query #1 + 10% filter 1710293 21 17.6 0.87 205 Skewed data, low filtering 10790019 21 96.8 2.47 39
6 Skewed data, high filtering 24 14 7.28 0.83 9
7 Many restrictions 3805941 21 128 7.65 17
8 Navigational attributes 823646 21 17.1 1.27 13
9 Navigational attributes + selective condition 811 21 15.8 1.17 1410 Open value ranges 2006 21 19.6 3.52 6
11 Hierarchy 1653981 21 17.6 0.97 18
12 Hierarchy + selective condition 55068 21 38.6 0.98 3913 Restricted key figures on 2 dimensions 1314964 1948 207 7.22 29
14 Query #14 + hierarchy 132564 1499 > 1000 1.27 > 78715 Calculated key figures (OLAP) 5321586 10 57.8 2.37 24
16 OR linked values 6212609 13 40.5 0.92 44
17 Non uniform data distribution 11016253 13 31.2 0.99 32
18 Selective line item 1724 1706 0.71 1.17 0.6
19 Non-selective line item 115481 68619 33.8 1.36 25
20 All together 3087692 468 87.7 4.42 20
1 3 5 7 9
11
13
15
17
19
0
20
40
60
80
100
120
140
Accelerator (sec) DB2 (sec)
18 million records in an InfoCube, 20 dedicated queries
© 2014 IBM Corporation19
Run queries up to 2000x faster
Business innovation with zEnterprise Analytics
Petrol, a strategic supplier of oil and other energy products is doing something they could never do before, increasing retail sales nearly 5% through reduced analytic query response times (99.8 % faster)
“The store employee enters what the customer is purchasing, and with the DB2 Analytics Accelerator appliance, the Cognos
BI and SPSS tools deliver information on complementary products in seconds.”
--A Chief Information officer--
© 2014 IBM Corporation20
Workload Assessment
1
Documentationand REXX procedure
Documentationand REXX procedure
Data package(mainly unload
data sets)
Data package(mainly unload
data sets)IBM lab
Database
IBM labDatabase
Pre-process andload
Pre-process andload
2 3
Quick Workload Test Tool
Quick Workload Test Tool
Report
Assessment
Report for a first assessment:� Acceleration potential for
• Queries• Estimated time• CP cost
Report for a first assessment:� Acceleration potential for
• Queries• Estimated time• CP cost
CustomerDatabase
CustomerDatabase
� Customer
• Collects information from dynamic statement cache, supported by step-by-step instruction and REXX script (small effort for customer)
• Upload compressed file (up to some MB) to IBM FTP server
� IBM
• Import data into local database
• Quick analysis based on known DB2 Analytics Accelerator capabilities
Key contact: Data Warehouse System z/Germany/IBM
20
© 2014 IBM Corporation21OMEGAMON XE for DB2 PE
Analyzeand
Report
DB2 Admin/OC
Manageand
Administer
Query Workload Tuner for z/OS
Compareand
Tune
Query Monitor for DB2
Monitorand
Identify
DB2 Analytics AcceleratorLoader
PerformanceLoad withoptions
IBM DB2 Tools and Accelerator Ecosystem
Is Acceleratorbeing under/over utilized?
How can Idetermine cost savings?
What queriesare being routed to theAccelerator?
How can I determinecost savings?
Do I have thebest set ofobjects for consideration?
Is there a wayto managefrom ISPF?
Can I load data from other sources?
How can Ideterminecost savings?
Can I improveon availabilitywhen loading?Can I tell how
current my data is?