Exadata Consolidation Success Story
Getting the kids to play nice with each other…
1
Presented by: Karl Arao
whoami
Karl Arao• Senior Technical Consultant @ Enkitec• Performance and Capacity Planning Enthusiast
7 years DBA experienceOracle ACE, OCP-DBA, RHCE, OakTableBlog: karlarao.wordpress.comWiki: karlarao.tiddlyspot.comTwitter: @karlarao
www.enkitec.com 2
www.enkitec.com 3
100+
3
Agenda
• Architecture • Tools and Methodology
– Simple consolidation scenario– Provisioning workflow and the worksheet
• War Stories
www.enkitec.com 4
General Architecture
www.enkitec.com 5
Primary Site Standby Site
Production
Test & Dev Disaster Recovery
Future Growth
General Architecture
www.enkitec.com 6
The Stats
Three Half Rack Exadata clusters with High Cap. drivesCluster #1
36 Dev/Test Databases
Cluster #211 Production Databases
Cluster #313 Dev/Test Databases6 Standby Databases
Still more databases to come…
www.enkitec.com 7
Why Consolidate?
Primary drivers for consolidation center around cost savings•Reduces Oracle software licensing•3rd party products such as backup agents, ETL tools, etc…•More efficient use of system resources•Soft Costs
– Floor space– Power & Cooling– Administration, Staffing Costs
(training, etc.)
www.enkitec.com 8
www.enkitec.com 9
7 Databases
A Simple Consolidation Example
www.enkitec.com 10
For example, the first row should read… Database ‘A’ requires 4 CPU’s and will run on nodes 1 and 2 (2 CPU’s each)
Let’s say we have the following databases to migrate on Exadata:
Cluster Level Utilization
A Simple Consolidation Example
www.enkitec.com 11
Let’s say we have the following databases to migrate on Exadata:
Per compute node Utilization
For example, the first row should read… Database ‘A’ requires 4 CPU’s and will run on nodes 1 and 2 (2 CPU’s each)
A Simple Consolidation Example
www.enkitec.com 12
Cluster Level
Utilization = 29.2%Per compute node Utilization
25% 42% 33% 17%
A Simple Consolidation Example
www.enkitec.com 13
Cluster Level
Utilization = 29.2%Per compute node Utilization
8% 83%83% 17% 8%
A Simple Consolidation Example
www.enkitec.com 14
• Gather Utilization Metrics (usage history)• Create Provisioning Plan• Implement Plan• Audit Your Implementation
Tools And Methodology
www.enkitec.com 15
Provisioning Worksheet
• Capacity Planning
• Communication Tool
• Hand off
www.enkitec.com 16
**Supplement to existing Exadata installation tools:• Site planning checklist• Configuration Worksheet• Exadata Configurator sheet• CheckIP• OneCommand
Utilization = Requirements / Capacity
Capacity
www.enkitec.com 17
2 = quarter rack4 = half rack8 = full rack
SPECint_rate2006http://goo.gl/doBI5
CPU_COUNT, threads, & cores
http://goo.gl/CunHN
96 to 144GB(frequency of the memory DIMMs
drops to 800 MHz from 1333 MHz)
Space will also depend on:•ASM redundancy•DATA/RECO allocation
http://goo.gl/I3fjn
Query Low (4x)Query High (6x)Archive Low (7x)
Archive High (12x)
Smart
Scans!
CPU Core Comparison
www.enkitec.com 18
Source
chip efficiency factor = source SPEC rating / Exadata SPEC rating = 16/26 = .6154 EXA cores requirement = source host cores * utilization * chip efficiency factor = 32 * .7 * .6154 = 13.78
* offload factor * .5 --------- 6.89
Sun Fire X4170 M2 [email protected]
Destination
how much of the source CPU cores
are being used
multiplier for equivalent database
machine cores
amount of CPU resources that will be offloaded to the
storage cells
The Perfect Storm(Peoplesoft HR)
www.enkitec.com 19
Month-end Processing+ Weekly Time Entry
+ SQL Plan Change------------------------------------
Uh-oh!
CPU Allocation
www.enkitec.com 20
DB Uniq Name DB Namenode1
node2
node3
node4
4 instance 5 instance 4 instance 3 instance
47% cpu used 75% cpu used 47% cpu used 18% cpu used
49% mem used 66% mem used 71% mem used 54% mem used
BIPRDDAL biprd P P
DBFSPRD DBFSPRD P P P P
HCMPRDDAL hcmprd P P
MTAPRD11DAL mtaprd11 P P
PAPRDDAL paprd P P
RMPRDDAL rmprd P P
dbm dbm F F F F
Fsprddal fsprd P P
= Preferred = Failover
www.enkitec.com 21
Load Map(our first stop…)
Users Complaint: HR time entry and OBIEE reports painfully slow…
www.enkitec.com 22
Top Activity - HCMPRD
www.enkitec.com 23
Instance Activity – HCMPRD2
HCMPRD Caged at 12 CPU’s
SQL Profile Installed to lock in good plan.
Problem: A single SQL stmt. overwhelming CPU resources.
Node 2
Memory Exhaustion(OBIEE)
“1 Report = 1 SQL query, right?”
WRONG!
www.enkitec.com 24
www.enkitec.com 25
Overlapping workloads of three databases across 3 nodes.
BIPRD, HCMPRD, and MTAPRD
Overlapping workloads of three databases across 3 nodes.
BIPRD, HCMPRD, and MTAPRD
Node 1
Node 2
Node 3
Node 4
www.enkitec.com 26
Node Layout Revisited…
www.enkitec.com 27
Notice what happens to CPU waits and the system load average when this report is run.
Notice what happens to CPU waits and the system load average when this report is run.
www.enkitec.com 28
PGA Memory SpikesPGA Memory Spikes
www.enkitec.com 29
www.enkitec.com 30
Storage Cell Saturation(OBIEE)
www.enkitec.com 31
www.enkitec.com 32
www.enkitec.com 33
I/O Intensive Workload
www.enkitec.com 34
Smart Scans as seen in Grid Control
www.enkitec.com 35
25 Sessions Doing Smart Scans
…as seen in gv$sql
www.enkitec.com 36
www.enkitec.com 37
Smart Scan in Action. The cells are scanning 1T but only returning 144G…***That’s on each of the highlighted row source below…
www.enkitec.com 38
The databases on other nodes see the contention as “System I/O”Without I/O resource management even critical processes are affected (CKPT, LGWR, …)
www.enkitec.com 39
Inter-database IORM Plan(only kicks in when needed)
I/O requests from critical processes like CKPT, LGWR, LMON get priority automatically.Without IORM I/O requests from these important processes receive the same priority as any other process.
*Side Benefit (automatic when IORM is enabled)
www.enkitec.com 40
IORM Plan Definition(on each storage cell)
Wrap up!
Provisioning Methodology & Tools– Workflow– Provisioning Spreadsheet
Success Stories– CPU resource management– Tuning and provisioning adjustments– I/O resource management
www.enkitec.com 41
www.enkitec.com 42
43
Fastest Growing Companies in Dallas
Contact Info…