Chapter 9: Monitoring and

Performance

Course 157: SQL Server 2008 Database Administration

Presented by Scott Whigham

1
p. 1
 What We’re Going to Cover
Overview

Monitoring
• Chapter • Understanding
Introduction and Using the
• Tools for DMVs
Monitoring • PerfMon
• SQL Server • Data Collector
Profiler

p. 2
 Chapter Introduction

• Chapter 9 is about performance and

exception monitoring your server
• We’ll start with the basics (using Profiler)
and end with new features in SQL Server
2008 (Data Collector)

p. 3
 What We’re Going to Cover
Overview

Monitoring
• Chapter • Understanding
Introduction and Using the
• Tools for DMVs
Monitoring • PerfMon
• SQL Server • Data Collector
Profiler

p. 4
 Tools of Performance Monitoring

• There are many built-in tools to monitor

performance and health:
– Activity Monitor and sp_who
• Covered in Chapter 5!
– SQL Server Profiler
• Detailed view of statements issued against instance
– Dynamic Management Views (DMVs)
• “Live” view of server health and activity
– Performance counters and objects (“PerfMon”)
• “Live” view of server health and activity
– Data Collector
• Historical view of server health and activity

p. 5
 Tools of Performance Monitoring

“But Mr. Scott, how do I know when to use each one?”

Perf Data
Profiler DMVs
Mon Coll.
Provides “live”, real-time reporting?

Can do trend analysis and historical reporting? Requires

manual load
Requires
manual load
Requires
manual load

Can view SQL issued?

Provides detail-level view or summary? Detail Depends… Summary Summary

Can be used to diagnose performance problems?

Level of impact on server Low None Negligible Low-Hi

Difficulty level 100-300 200-400 200-300 300-500

p. 6
 In the next video…

• Let’s play with Profiler!

p. 7
Chapter 9: Monitoring and
Performance

Course 157: SQL Server 2008 Database Administration

Presented by Scott Whigham

8
p. 8
 What We’re Going to Cover
Overview

Monitoring
• Chapter • Understanding
Introduction and Using the
• Tools for DMVs
Monitoring • PerfMon
• SQL Server • Data Collector
Profiler

p. 9
 SQL Server Profiler

• SQL Server Profiler is a tool for

monitoring real-time queries issued
against your server
– “Old Faithful” - been around for years
– Requires sysadmin or ALTER TRACE server
permission

p. 10
SQL Server Profiler

p. 11
 SQL Server Profiler

• DBAs use traces to monitor statements

sent to SQL Server
– We’ve already seen how SQL Server uses “default
traces” in Chapter 7
• Running traces will have an impact on
server performance
– How much? It depends on how much you trace

p. 12
 SQL Server Profiler

• SQL Server Profiler is the GUI for

defining, viewing, and replaying traces
– It does not have to be running to create a trace
– Many DBAs create their traces via Profiler and
then script out to jobs

p. 13
 SQL Server Profiler

• Running Profiler on a “live” system adds

to an already stressful situation
– Particularly if run interactively on the server
• If you must run Profiler against a “live”
server, run it remotely
• Adds network overhead but this is almost never an
issue today

p. 14
 SQL Server Profiler

• Trace data can be stored:

– In a .trc file (fastest)
– In a SQL Server table
• .trc files can be configured to rollover to
a new file after a certain threshold
• Can load .trc file contents into SQL Server
table after trace for maximum flexibility!

p. 15
 SQL Server Profiler

• After trace is complete, you can export

results

p. 16
 SQL Server Profiler

• Tips on working with Profiler

– Traces can have filters to help minimize the noise
• The better you filter, the less of an impact you put on
the server
– Being able to write queries against the trace
results can help you identify many common trends
and problems
• This requires you load the trace data into a table!

p. 17
 In the next video…

• Let’s play with Profiler!

p. 18
Chapter 9: Monitoring and
Performance

Course 157: SQL Server 2008 Database Administration

Presented by Scott Whigham

19
p. 19
 What We’re Going to Cover
Overview

Monitoring
• Chapter • Understanding
Introduction and Using the
• Tools for DMVs
Monitoring • PerfMon
• SQL Server • Data Collector
Profiler

p. 20
 Dynamic Management Views

• DBAs use Dynamic Management Views

(“DMVs”) and Dynamic Management
Functions (“DMFs”) to peek under the
covers
– In the sys schema and begin with dm_
– Used to get up-to-the-second performance
information during critical time
– Often need to JOIN many together to come up
with desired results

p. 21
 Tools of Performance Monitoring

Reminder…
Perf Data
Profiler DMVs
Mon Coll.
Provides “live”, real-time reporting?

Can do trend analysis and historical reporting? Requires

manual load
Requires
manual load
Requires
manual load

Can view SQL issued?

Provides detail-level view or summary? Detail Depends… Summary Summary

Can be used to diagnose performance problems?

Level of impact on server Low None Negligible Low-Hi

Difficulty level 100-300 300-500 200-300 300-500

p. 22
 Dynamic Management Views

• There are too many DMxs to cover each

one so we’ll cover concepts and core
DMxs
– Over 130 DMVs and DMFs in SQL Server 2008!
• It is important for DBAs to maintain a
script library

p. 23
 Dynamic Management Views

• The easiest way to get to know the

DMVs/DMFs is to categorize them
– Below are the more popular categories for
performance monitoring
Category DMx prefix
Database sys.dm_db_
Execution/Sessions sys.dm_exec_
IO sys.dm_io_
OS sys.dm_os_
Transactions sys.dm_tran_

p. 24
 Dynamic Management Views

• Intellisense can help you greatly!

p. 25
 Dynamic Management Views

• Object Explorer can also help

p. 26
 DMV and DMF Query

• You can also query system views for full

list

SELECT name, type_desc

FROM sys.system_objects
WHERE schema_id=4
AND name LIKE 'dm_%‘
ORDER BY name

p. 27
DMV and DMF Query Results

p. 28
 Database Category

• Database category:
– 16+ DMVs and DMFs

SELECT name, type_desc

FROM sys.system_objects
WHERE schema_id=4
AND name LIKE 'dm_db%'

p. 29
 In the next video…

• We’ll take a look at a few categories and

examples

p. 30
Chapter 9: Monitoring and
Performance

Course 157: SQL Server 2008 Database Administration

Presented by Scott Whigham

31
p. 31
 What We’re Going to Cover
Overview

Monitoring
• Chapter • Understanding
Introduction and Using the
• Tools for DMVs
Monitoring • PerfMon
• SQL Server • Data Collector
Profiler

p. 32
 Dynamic Management Views

• Reminder: the more popular categories

for performance monitoring
Category DMx prefix
Database sys.dm_db_
Execution/Sessions sys.dm_exec_
IO sys.dm_io_
OS sys.dm_os_
Transactions sys.dm_tran_

p. 33
 Database Category

• Database category:
– 16+ DMVs and DMFs

SELECT name, type_desc

FROM sys.system_objects
WHERE schema_id=4
AND name LIKE 'dm_db%‘
ORDER BY name

p. 34
 Database Category

• From a performance standpoint, you will

spend a lot of time in the Database
category
– Primarily looking at table, index, and partition
information
• Size, fragmentation
• Missing indexes
• Index use

p. 35
 Database Category

• Common index-related DMVs in the

Database Category
– sys.dm_db_missing_index_details
returns information about indexes that could
potentially help speed up the system
– sys.dm_db_index_usage_stats can tell
you whether an index is being used and, if so, in
what way

p. 36
 Database Category

• Common index-related DMFs in the

Database Category
– sys.dm_db_index_physical_stats is
one way to identify fragmentation (see Chapter
8!)
– sys.dm_db_index_operational_stats
is similar to the DMV sys.dm_db_index_usage_stats
yet only includes actual execution events instead
of plan generation events

p. 37
 In the next video…

• We’ll take a look at writing queries using

the DMVs and DMFs in the Database
category

p. 38
Chapter 9: Monitoring and
Performance

Course 157: SQL Server 2008 Database Administration

Presented by Scott Whigham

39
p. 39
 What We’re Going to Cover
Overview

Monitoring
• Chapter • Understanding
Introduction and Using the
• Tools for DMVs
Monitoring • PerfMon
• SQL Server • Data Collector
Profiler

p. 40
 Dynamic Management Views

• Let’s look at the other categories now…

Category DMx prefix

Database sys.dm_db_
Execution/Sessions sys.dm_exec_
IO sys.dm_io_
OS sys.dm_os_
Transactions sys.dm_tran_

p. 41
 Execution/Session Category

• Execution/Session category:
– 20+ DMVs and DMFs

SELECT name, type_desc

FROM sys.system_objects
WHERE schema_id=4
AND name LIKE 'dm_exec%'

p. 42
 Execution/Session Category

• The Execution/Session category will help

you spy on people
– That’s what you really want to do, isn’t it?
– Can also view specific query information
• Plans, text, stats, caching information

p. 43
 Execution/Session Category

• There is at least one DMF in this category

that you definitely want to know
– sys.dm_exec_sql_text allows you to
see the query text a user submits
• By itself, it is not that useful but combined with other
Execution/Session DMVs it is very powerful
• Accepts one parameter - the memory location of the
query plan -
– Populated via a column found in other sys.dm_exec_
DMVs

p. 44
 Execution/Session Category

• A common DMV in this category is:

– sys.dm_exec_query_stats allows you to see
metrics about queries submitted to server
• Reads, writes, waits, execution count, etc
• plan_handle column is the column to pass to
sys.dm_exec_sql_text to return SQL

p. 45
 Return 50 most-run queries

SELECT TOP(50) sql.text

, qs.last_execution_time
, qs.execution_count
, qs.last_elapsed_/1000000.0 as
ElapsedTimeInSecondstime
FROM sys.dm_exec_query_stats qs
CROSS APPLY
sys.dm_exec_sql_text(qs.sql_handle) AS
sql
ORDER BY qs.execution_count DESC

p. 46
Return 50 most-run queries

p. 47
 Execution/Session Category

• Other common DMVs in this category are

– sys.dm_exec_procedure_stats same as
query_stats yet only returns stored procedure
info
– sys.dm_exec_cached_plans gives you info on
the cached query plans
• Is it compiled or adhoc?
• How many times has it been used?

p. 48
 Execution/Session Category

• Spying on users is easy!

– sys.dm_exec_sessions is one row for each
authenticated session

-- Returns all non-internal sessions

SELECT * FROM sys.dm_exec_sessions
WHERE session_id>50

p. 49
 In the next video…

• We’ll write several queries and see the

Execution/Session category in action

p. 50
Chapter 9: Monitoring and
Performance

Course 157: SQL Server 2008 Database Administration

Presented by Scott Whigham

51
p. 51
 What We’re Going to Cover
Overview

Monitoring
• Chapter • Understanding
Introduction and Using the
• Tools for DMVs
Monitoring • PerfMon
• SQL Server • Data Collector
Profiler

p. 52
 Dynamic Management Views

• Let’s look at the IO and OS categories

now to help diagnose disk pressure
Category DMx prefix
Database sys.dm_db_
Execution/Sessions sys.dm_exec_
IO sys.dm_io_
OS sys.dm_os_
Transactions sys.dm_tran_

p. 53
 IO Category

• IO category:
– 4+ DMVs and DMFs

SELECT name, type_desc

FROM sys.system_objects
WHERE schema_id=4
AND name LIKE 'dm_io_%‘
ORDER BY name

p. 54
 IO Category

• The most common item in this category

is
– sys.dm_io_virtual_file_stats()
returns information about usage of the database
files
• Reads, writes, IO waits, size
• Accepts two parameters:
– database_id - use NULL for all databases
– file_id - the file_id from sys.database_files; use NULL for
all files

p. 55
 IO Category

• sys.dm_io_virtual_file_stats
returns several columns:
Column Description
sample_ms Milliseconds since server was restarted
num_of_reads Physical reads of objects in database file
num_of_bytes_read Large textual-based columns will affect this greatly
io_stall_read_ms Total time a user process waited for IO to this file
num_of_writes, Database write activity
num_of_bytes_written,
io_stall_write_ms
io_stall Total amount of waits reading /writing to this file

p. 56
 IO Category

• Here are some things to monitor with

dm_io_virtual_file_stats()
– Look at total reads+writes to find “most active”
– Look at io_stall to identify files that might
benefit from new disks
• All databases have stalls; the important metric is “What
percentage of the uptime of the server am I having
stalls?”
• Compare read/write stalls on log and data file to
identify which could benefit from new/faster disks

p. 57
 IO DMF

-- View IO stats for all DBs on instance:

SELECT DB_NAME(mf.database_id) AS DB
, mf.physical_name, mf.type_desc
, vfs.*
FROM
sys.dm_io_virtual_file_stats(null,null)vfs
JOIN sys.master_files mf
ON mf.database_id = vfs.database_id
AND mf.file_id = vfs.file_id
ORDER BY DB_NAME(mf.database_id)

p. 58
IO DMF

p. 59
 IO DMF

-- Aggregate total IO since last restart:

SELECT
DB_NAME(vfs.database_id) AS database_name,
CAST(SUM(vfs.num_of_bytes_read +
vfs.num_of_bytes_written) / 1048576.
AS DECIMAL(12, 2)) AS IO_MB
FROM sys.dm_io_virtual_file_stats(NULL, NULL) vfs
GROUP BY vfs.database_id
ORDER BY IO_MB DESC

p. 60
 OS Category

• The OS category is one of the largest

– 36+ DMVs (no DMFs!)

SELECT name, type_desc

FROM sys.system_objects
WHERE schema_id=4
AND name LIKE 'dm_os_%‘
ORDER BY name

p. 61
 OS Category

• You can use this category to find waiting

tasks:
– sys.dm_os_wait_stats - aggregated info on
dozens and dozens of wait types
– sys.dm_os_waiting_tasks - detail info on
waits for a specific resource
• Hint: use dm_os_wait_stats to identify the areas of
contention and then dm_os_waiting_tasks to view
task information

p. 62
 OS Category

• These two DMVs are accumulative

– Their statistics are the totals since SQL Server
service has been started
• To track effectively over time, it is
necessary to reset periodically
– Common to store in a table for historical
perspective
– DBCC SQLPERF(‘<dmv_name>’, CLEAR)

p. 63
 OS Category

-- Current wait info:

SELECT wait_type, wait_time_ms / 1000. as wait_time_s
FROM sys.dm_os_wait_stats
ORDER BY wait_time_ms DESC

-- Clear out stats:

DBCC SQLPERF ('sys.dm_os_wait_stats', CLEAR)

-- Run this at a later time to view waits since CLEAR

SELECT wait_type, wait_time_ms / 1000. as wait_time_s
FROM sys.dm_os_wait_stats
ORDER BY wait_time_ms DESC

p. 64
OS Category

p. 65
 OS Category

• To identify disk pressure, you can look in

sys.dm_os_wait_stats for specific wait
types
– Order by wait_time_ms DESC to identify most often
waits
– If anyone of these are in the top two positions, you
have disk I/O pressure:
• ASYNCH_IO_COMPLETION
• IO_COMPLETION
• LOGMGR
• WRITELOG
• PAGEIOLATCH_x

p. 66
 In the next video…

• Let’s look further at the OS category and

others

p. 67
Chapter 9: Monitoring and
Performance

Course 157: SQL Server 2008 Database Administration

Presented by Scott Whigham

68
p. 68
 What We’re Going to Cover
Overview

Monitoring
• Chapter • Understanding
Introduction and Using the
• Tools for DMVs
Monitoring • PerfMon
• SQL Server • Data Collector
Profiler

p. 69
 OS Category

• The OS category has lots of info

– sys.dm_os_sys_info - info such as number
of CPUs, start time of service
– sys.dm_os_tasks - lists each running task
– sys.dm_os_loaded_modules - lists each
loaded DLL that SQL Server is using
• Cryptography Manager
• Kerberos Security Package
• Microsoft SQL Server Native Client 10.0
• SQL Server Windows 95 Lite SCM

p. 70
 OS Category

• Many OS category DMVs can help

diagnose memory pressure
– Performance Counters can help too
– We’ll cover Perf Counters in an upcoming section
of this chapter

p. 71
 OS Category

• The main areas for memory pressure are:

– Database page cache (used to cache table/index)
– Query Workspace memory (for Hash and Sort ops)
– Plan cache
– There are others…

p. 72
 OS Category

-- Find out what has the most entries

-- in memory
SELECT TOP (20)[Name], COUNT(*) AS 'Count'
FROM sys.dm_os_memory_cache_entries
GROUP BY [Name]
ORDER BY COUNT(*) DESC;

p. 73
OS Category

p. 74
 OS Category

-- Top 10 consumers of Buffer Pool

SELECT TOP 10 [type]
, SUM(single_pages_kb) AS memory_in_kb
FROM sys.dm_os_memory_clerks
GROUP BY type
ORDER BY SUM(single_pages_kb) DESC;

p. 75
OS Category

p. 76
 OS Category

• One area that many servers have trouble

with is the TokenAndPermUserStore cache
– An internal security system that generates security
tokens for sessions
– Particularly troublesome area for DBs with lots of ad-
hoc SQL
– The problem can be that the
TokenAndPermUserStore cache takes up so
much memory and is so frequently accessed that
when there are a high number of items in the cache,
access takes a long time

p. 77
 OS Category

1. Queries for diagnosing

TokenAndPermUserStore cache
issues:

-- What is the size? The larger, the slower!

SELECT SUM(single_pages_kb + multi_pages_kb)
AS ‘CurrentSizeOfTokenCache(kb)’
FROM sys.dm_os_memory_clerks
WHERE name = 'TokenAndPermUserStore'

p. 78
 OS Category

2. Queries for diagnosing

TokenAndPermUserStore cache
issues:

-- How many entries in the

-- TokenAndPermUserStore?
SELECT TOP (20)[Name], COUNT(*) AS 'Count'
FROM sys.dm_os_memory_cache_entries
GROUP BY [Name]
ORDER BY COUNT(*) DESC;

p. 79
Scary!

p. 80
 OS Category

• There are multiple ways to prevent

TokenAndPermUserStore issues
1. Minimize ad-hoc SQL
2. Log in as a sysadmin (no tokens need to be
generated)
3. Limit the size of the cache via sp_configure
using
• access check bucket count
• access check cache quota
• http://support.microsoft.com/kb/955644

p. 81
 OS Category

• There is one way to fix a “live”

TokenAndPermUserStore issue
– Free the system cache

-- WARNING: Do not use on production without

-- a full understanding!
-- After running, there will not be an entry
-- in sys.dm_os_memory_clerks for TAPUS!
DBCC FREESYSTEMCACHE('TokenAndPermUserStore')

p. 82
 In the next video…

• Let’s look at blocking and locking…

p. 92
Chapter 9: Monitoring and
Performance

Course 157: SQL Server 2008 Database Administration

Presented by Scott Whigham

93
p. 93
 What We’re Going to Cover
Overview

Monitoring
• Chapter • Understanding
Introduction and Using the
• Tools for DMVs
Monitoring • PerfMon
• SQL Server • Data Collector
Profiler

p. 94
 Performance Monitor

• To maximize coverage, you want to use

both DMVs/DMFs along with
Performance Monitor
– a.k.a. PerfMon and System Monitor
• PerfMon gathers stats about SQL Server
as well as the rest of the environment
– Disk subsystems, non-SQL resource usage

p. 95
 Performance Monitor

• PerfMon uses objects and counters

– Objects are for categorization
– Objects contain one or more counters
• When you install SQL Server, it installs a
set of instance-specific objects

p. 96
 Performance Monitor

• You can view the SQL Server objects and

counters using the DMV
sys.dm_os_performance_counters
– Careful - more than 1,100 rows

p. 97
 Performance Monitor

• A few things to know about

sys.dm_os_performance_counters
– Only displays SQL Server instance’s counters
– Any counter that has “/sec” in the name is cumulative
since time SQL Server service was started
• Transactions/sec
• Bulk copy rows/sec
– To accurately read these values, compare
intervals
• Compare 0800 reading to 0900 reading, for
example

p. 98
 Performance Monitor

• Example query to return all

deprecated features usage
– Many feature usages are actually SQL Server
internals and tools!
SELECT *
FROM sys.dm_os_performance_counters pc
WHERE object_name
LIKE '%Deprecated Features%'
AND cntr_value > 0
ORDER BY counter_name

p. 99
Performance Monitor

p. 100
 Performance Monitor

• Querying for specific counters allows ad-

hoc views of point-in-time information
– But no graphs, charts, pretty colors
• For trend analysis, you have a few
choices:
– Store the query results in a table for later review
– Use PerfMon to store logs for later analysis

p. 101
 Performance Monitor

• Some DBAs will store certain values in a

table for later querying or for historical
perspective
– Watch the rest of the chapter prior to attempting
this!
-- Create a new table (”PerfCounters”)
-- and store current counters
SELECT *
INTO PerfCounters
FROM sys.dm_os_performance_counters

p. 102
 Performance Monitor

• All DBAs will use PerfMon

p. 103
 In the next video…

• Let’s take a look at specific objects and

counters

p. 104
Chapter 9: Monitoring and
Performance

Course 157: SQL Server 2008 Database Administration

Presented by Scott Whigham

105
p. 105
 What We’re Going to Cover
Overview

Monitoring
• Chapter • Understanding
Introduction and Using the
• Tools for DMVs
Monitoring • PerfMon
• SQL Server • Data Collector
Profiler

p. 106
 Performance Monitor

• There are a few specific areas you will

want to monitor
– Sometimes you can find that “Aha!” moment
using ad-hoc monitoring
– Sometimes you can only find that “Aha!” moment
using trend analysis

p. 107
 Performance Monitor

• A few topics before we go further:

– PerfMon, counters with “/sec” in the name are
not cumulative
• Makes it easy to do ad-hoc monitoring
– PerfMon’s GUI defaults to taking measurements
every one second
• Too often for trend analysis
• 30-seconds is generally “good enough” for ad-hoc
troubleshooting

p. 108
 Performance Monitor

• A few areas DBAs want to monitor:

– Disk usage
– CPU usage
– Memory usage
• Let’s look at each!

p. 109
 Performance Monitor

• A few areas DBAs want to monitor:

– Disk usage
– CPU usage
– Memory usage

p. 110
 Disk Usage

• Disk usage monitoring

– Disk I/O is commonly a performance stealer
– While SQL Server manages the “when” and “how
often” of disk I/O, it is the operating system that
performs I/O
– When monitoring disk activity, you need to
monitor both SQL Server’s disk activity and the
disk activity of the server as a whole

p. 111
 Disk Usage

• Disk I/O cannot be isolated as a metric

– Example: You are monitoring disk I/O and notice
that you have waits for disk reads/writes. Does
this always mean that you need faster/more
disks?
• No! Could be that you are paging constantly and
memory is the real bottleneck

p. 112
 Disk Usage

• Important system objects and counters:

Object Counter When to worry
PhysicalDisk % Disk Time Consistently > 90%
PhysicalDisk Current Disk Queue Particularly helpful when “% Disk Time” is
Length high; allows to see how many I/O
operations are waiting. Start worrying
when this value is more than 1.5x the
number of spindles
PhysicalDisk Avg. Disk Queue Length In many configs, a disk has 1 spindle. Start
worrying when this value is more than 1.5x
the number of spindles

Memory Page faults/sec “Some” page faults are normal; look for
consistently high numbers

p. 113
 Disk Usage

• PhysicalDisk: % Disk Time

– Tells us how “busy” your array is
– With this counter, you want to worry about
sustained activity, not bursts
– A % Disk Time of more than 75% for several
continuous minutes occurring several times a day
can likely be resolved:
• By adding more disks to the array
• By using faster drives

p. 114
 Disk Usage

• PhysicalDisk: % Disk Time (cont.)

– When you do notice this is high, start looking at
the PhsyicalDisk: Current Disk Queue Length
• How many disks are in my array?
– Tells you how many spindles you have
• How many I/O tasks are waiting?
– If the number of waiting tasks is 1.5 or higher per
spindle, you need better I/O performance!

p. 115
 Disk Usage

• PhysicalDisk: Avg. Disk Queue Length

– Monitor this over time
• Business hours
• 24 hour period
• Peak times
– If you find that it exceeds more than 1.5 waiting
tasks per spindle for continuous periods, increase
disk performance

p. 116
 Disk Usage

• Important SQL Server objects & counters

to help diagnose I/O trouble
– Object:
• SQL Server: Buffer Manager
– Counters:
• Page reads/sec
• Page writes/sec

p. 117
 Disk Usage

• Page reads/sec and Page writes/sec

measure physical reads, not logical reads
– Not logical reads!
– Object:
• SQL Server: Buffer Manager
– Counters:
• Page reads/sec
• Page writes/sec

p. 118
 One Caveat

• High disk usage is not always a sign that

you need more/faster disks
– Sometimes poorly optimized queries can
masquerade as disk/memory/CPU issues
– Proper indexing and better queries can help
eliminate slow queries as the culprit

p. 119
 A Final Caveat

• PerfMon is great for monitoring disks

directly attached to the server
– Good but not great for monitoring SANs
– Check with your SAN vendor for suggestions on
monitoring disk activity

p. 120
 In the next video…

• Let’s play with PerfMon and look at Disk,

CPU, and memory counters

p. 121
Chapter 9: Monitoring and
Performance

Course 157: SQL Server 2008 Database Administration

Presented by Scott Whigham

122
p. 122
 What We’re Going to Cover
Overview

Monitoring
• Chapter • Understanding
Introduction and Using the
• Tools for DMVs
Monitoring • PerfMon
• SQL Server • Data Collector
Profiler

p. 123
 Data Collector

• While it is fun and educational to learn

the various DMVs and DMFs along with
the relationships between various system
views, it is a huge time sink
– Pro tip: download the SQL Server 2008 System
Views map
– http://learnitfirst.com/go.aspx?id=5GR5

p. 124
 Data Collector

• How to be an unsuccessful DBA:

– Try to remember all the DMVs, DMFs, and
Performance Counters
– Perform all of your performance data collection by
running 100+ line queries daily
• Bonus points if you are the only person who knows
where on the network these queries/files are located
– Spend 20 hours a week writing and running DMV-
based queries

p. 125
 Data Collector

• How to be an successful DBA:

– Learn the most important DMVs, DMFs and
Performance Counters for your situation
• Every job is different and every database is different
• OLTP databases need different metrics than OLAP
• DBAs of a “single database only” server are often concerned
with different things than DBAs of “20 database” servers
– Make your reporting easy to digest and easy to run
• No more 20 hours a week manually writing/running DMV-
based queries!

p. 126
 Data Collector

• Life as a DBA has traditionally been spent

doing a lot of manual collection of
performance data
1. Create a tables to store performance data
2. Write the queries to collect the data
3. Create a job to store query results in tables
4. Write queries against tables to determine
trends, etc

p. 127
 Data Collector

• This was how things were done for more

than a decade
– As recent as SQL Server 2000…
– SQL Server 2005 debuted “Custom Reports” to
help us
• Allowed integration of SSMS with Reporting Services to
create professional quality reports based on our
queries
• “SQL Server 2005 Performance Dashboard” reports

p. 128
 Data Collector

• SQL Server 2008 includes the Data

Collector to save you time
– Creates a Management Data Warehouse (MDW)
that stores critical tuning, planning, and activity
history
– Used daily by DBAs to do capacity planning,
mitigate risks, free up time otherwise spent trying
to write complex DMV/DMF queries!

p. 129
 Data Collector

• The Data Collector is turned off by

default
– Enabling it is not a trivial decision…
– Performance data collection will impact your
system
• It’s the Heisenberg uncertainty principle in action!
• How much of an impact depends on what you collect
and how often

p. 130
 Data Collector

• A disclaimer:
– Please test, test, test this prior to deploying in a
production environment
– This is version 1.0 software
– Easy to “over collect”
• 30GB of data collection in three days!

p. 131
 Data Collector

• Q& for Data Collector

– “What type of information can DC collect?”
• Traces, PerfMon counters, SQL queries
– “Can I use DC to collect SQL 2005/2000 servers?”
• No, SQL Server 2008 only
– “Is this another one of those ‘Enterprise Edition
only’ features?”
• No - works great in Standard and Workgroup Editions
too!

p. 132
 Data Collector

• Q& for Data Collector (cont.)

– “How much of an impact will this have on my
server?”
• Microsoft says “5% of CPU and 250-350MB” for the
built-in collection sets. Custom collection sets that you
create may consume more
– More on this later…

p. 133
 In the next video…

• Let’s learn a bit more about the Data

Collector

p. 134
Chapter 9: Monitoring and
Performance

Course 157: SQL Server 2008 Database Administration

Presented by Scott Whigham

135
p. 135
 What We’re Going to Cover
Overview

Monitoring
• Chapter • Understanding
Introduction and Using the
• Tools for DMVs
Monitoring • PerfMon
• SQL Server • Data Collector
Profiler

p. 136
 Data Collector

• The Data Collector (“DC”) is built into

SSMS

p. 137
 Data Collector

• You can enable the DC via the Configure

Management Data Warehouse Wizard

p. 138
 Data Collector

• Configuring the DC is a two-step process:

1. Creating the Management Data Warehouse (the
“MDW”)
2. Setting up the actual collection of metrics

p. 139
 Data Collector

• Configuring the DC is a two-step process:

1. Creating the Management Data Warehouse (the
“MDW”)
2. Setting up the actual collection of metrics

p. 140
 Data Collector

• The MDW is a relational data warehouse

to store all of your metrics for later
analysis

p. 141
 Data Collector

• When you create the MDW database,

you define the name and physical file
location
• Placing the MDW database on either a
dedicated server or dedicated disks will
minimize performance impact to
monitored servers

p. 142
 Data Collector

• The MDW data can be shared with non-

sysadmins
– All sysadmins have access
– Often want network admins to have access to
performance data without having access to
specific databases

p. 143
 Data Collector

• There are three databases roles you can

assign:

p. 144
 Data Collector

• Once you decide security, you are ready

to build the MDW

p. 145
 In the next video…

• Configuring the DC is a two-step process:

1. Creating the Management Data Warehouse (the
“MDW”)
2. Setting up the actual collection of metrics

p. 146
Chapter 9: Monitoring and
Performance

Course 157: SQL Server 2008 Database Administration

Presented by Scott Whigham

147
p. 147
 What We’re Going to Cover
Overview

Monitoring
• Chapter • Understanding
Introduction and Using the
• Tools for DMVs
Monitoring • PerfMon
• SQL Server • Data Collector
Profiler

p. 148
 Data Collector

• Configuring the DC is a two-step process:

1. Creating the Management Data Warehouse (the
“MDW”)
2. Setting up the actual collection of metrics

p. 149
 Data Collector

• Back through the wizard…

p. 150
 Data Collector

• The Cache directory is local storage on

the monitored machine
– Remember that this can be run from a central server

p. 151
 Data Collector

• Data collector will create SQL Server jobs

to perform data collection
• The job has two main components:
1. Collect the info
2. Upload the info to the MDW
• Where you collect the info locally is the cache
directory
– Generally the %temp% folder

p. 152
 Data Collector

• Ready, set, go!

p. 153
 Data Collector

• What it should look like

p. 154
 Data Collector

• Now comes the work…

– You can use the System Data Collection Sets
– You can create your own collection sets
• There is no graphical way to create your own
collection sets
– Must write Transact-SQL

p. 155
 In the next video…

• We’ll take a look at enabling the Data

Collector and viewing the results

p. 156