SQL Server

Report
Garrett Edmondson
Data Warehouse Architect
Blue Granite Inc.
[email protected]
http://garrettedmondson.wordpress.com/
DW vs. OLTP
 OLTP
 Seek Centric (Radom Read/Writes) IOPs
 Volatile Data
 Index Heavy
 Many Heap Tables
 High Concurrency
• Data Warehouse
– Scan Centric (sequential reads/writes) MB/sec
– Nonvolatile Data – nightly loads
– Index – Light
• Few covering Clustered Indexes
– Low Concurrency
Traditional SQL DW
Architecture
Shared Infrastructure
Data Warehouse Speed = THROUGHPUT
from storage to CPU
Dedicated Network
Bandwidth
Enterprise Shared
SAN Storage
DW
The DW Trinity
• Star Schema (Fact & Dimension)
– The Data Warehouse Toolkit: The Complete Guide
to Dimensional Modeling (Second Edition)
• Balanced Architecture
– Parallel Data Warehouse
• ColumnStore Indexes - SQL 2012
SQL Server Columnstore Index FAQ
– “Batch” Mode
– Page Compression
CPU Feed Rate
FC
HBA
SQL Server
Read Ahead Rate
A
B
A
B
HBA Port Rate
A
A
B
STORAGE
CONTROLLER
CACHE
FC
HBA
DISK
FC SWITCH
CPU CORES
WINDOWS
SQL SERVER
CACHE
SERVER
Potential Performance Bottlenecks
A
B
Switch Port Rate
LUN
DISK
B
SP Port Rate
DISK
DISK
LUN
LUN Read Rate
Disk Feed Rate
Balanced System IO Stack
Tune every IO Interface !
CPU
Socket
(4 Core)
CPU Socket
(4 Core)
SMP Scale-Up for Data Growth!!!
• Complex
• Costly
• Obsolete Hardware
CPU Maximum Consumption Rate
(MCR)
• Theoretical Max (MCR)
– From Cache Query
• Set STATISTICS IO and STATISTICS TIME to ON
• MAXDOP = 4
• ( [Logical reads] / [CPU time in seconds] ) * 8KB / 1024 =
MCR per core
• Benchmark (BCR)
– From Disk - Cache
•
•
•
•
DBCC dropcleanbuffers
Set STATISTICS IO and STATISTICS TIME to ON
MAXDOP 8+
( [Logical reads] / [CPU time in seconds] ) * 8KB / 1024
MB/sec
• Test Storage Throughput in MB/sec
– SQLIO
– Sequential Reads and Writes Important
Establish real rather than rated, performance
metrics for the key hardware components of the
Fast Track reference architecture!!!!
LUN Configuration
Fast Track Data Striping
Storage Enclosure
ARY01D1v01
DB1-1.ndf
ARY02D1v03
DB1-3.ndf
ARY03D1v05
DB1-5.ndf
ARY04D1v07
ARY05v09
DB1-7.ndf
DB1.ldf
ARY01D2v02
DB1-2.ndf
ARY02D2v04
DB1-4.ndf
ARY03D2v06
DB1-6.ndf
ARY04D2v08
DB1-8.ndf
Evaluating Page Fragmentation
Average Fragment Size in Pages – This metric is a reasonable measure of contiguous page
allocations for a table.
Value should be >= 400 for optimal performance
select db_name(ps.database_id) as database_name
,object_name(ps.object_id) as table_name
,ps.index_id
,i.name
,cast (ps.avg_fragment_size_in_pages as int) as [Avg Fragment Size In Pages]
,ps.fragment_count as [Fragment Count]
,ps.page_count
,(ps.page_count * 8)/1024/1024 as [Size in GB]
from sys.dm_db_index_physical_stats (DB_ID()
--NULL for all DBs else run in context of DB
, OBJECT_ID(‘dbo.lineitem’), 1, NULL, ‘SAMPLED’) AS ps --DETAILED, SAMPLED, NULL = LIMITED
inner join sys.indexes AS i
on (ps.object_id = i.object_id AND ps.index_id = i.index_id)
where ps.database_id = db_id()
and ps.index_level = 0
TRACE FLAG –E !!!!!!
Trace FLAG -T1117
(c) 2011 Microsoft. All rights reserved.
SQL Server Configurations
• Sequential scan performance starts with
database creation and extent allocation
• Recall that the –E startup option is used
– Allocate 64 extents at a time (4MB)
• Pre-allocation of user databases is strongly
recommended
• Autogrow should be avoided if possible
– If used, always use 4MB increments
Storage Layout Best Practices for SQL
Server
• Create a SQL data file per LUN, for every filegroup
• TempDB filegroups share same LUNs as other databases
• Log on separate disks, within each enclosure
– Striped using SQL Striping
– Log may share these LUNs with load files, backup targets
Storage Layout Best Practices for SQL Server
TempDB
Stage
Database
Permanant_DB
LUN 1
LUN 2
LUN16
LUN 3
Permanent FG
Permanent_1.ndf
Permanent_2.ndf
Permanent_16.ndf
Permanent_3.ndf
Stage FG
Stage_1.ndf
Stage_2.ndf
Stage_3.ndf
Stage_16.ndf
Local Drive 1
TempDB.mdf (25GB)
TempDB_02.ndf (25GB)
TempDB_03ndf (25GB)
TempDB_16.ndf (25GB)
Log LUN 1
Permanent DB
Log
Stage DB Log
Techniques to Maximize Scan
Throughput
•
–E startup parameter (2MB Extents and not mixed extents)
•
Minimize use of NonClustered indexes on Fact Tables
•
Load techniques to avoid fragmentation
– Load in Clustered Index order (e.g. date) when possible
•
Index Creation always MAXDOP 1, SORT_IN_TEMPDB
•
Isolate volatile tables in separate filegroup
•
Isolate staging tables in separate filegroup or DB
•
Periodic maintenance
•
Turn on SQL Server Compression
Conventional data loads lead to
fragmentation
• Bulk Inserts into Clustered Index using a
moderate ‘batchsize’ parameter
– Each ‘batch’ is sorted independently… causes
fragmentation
• Overlapping batches lead to page splits
1:31
1:32
1:36
1:33
1:32
1:34
1:37
Key Order of Index
1:35
1:33
1:38
1:34
1:39
1:35
1:40
Best Practices for loading
•
•
•
Use a heap
– Practical if queries need to scan whole partitions
or…Use a batchsize = 0
– Fine if no parallelism is needed during load
or…Use a Two-Step Load
1. Load to a Staging Table (heap)
2. INSERT-SELECT from Staging Table into Target CI
Resulting rows are not fragmented
Can use Parallelism in step 1 – essential for large data volumes
Other fragmentation best practices
• Avoid Autogrow of filegroups
– Pre-allocate filegroups to desired long-term size
– Manually grow in large increments when necessary
• Keep volatile tables in a separate filegroup
– Tables that are frequently rebuilt or loaded in small
increments
• If historical partitions are loaded in parallel,
consider separate filegroups for separate
partitions to avoid extent fragmentation
Columnstore Indexes
storage = Segments + Dictionaries
CPU = Batch Mode
(c) 2011 Microsoft. All rights reserved.
Row Storage Layout
Customers Table
ID
Name
Address
City
State
1
Bob
…
…
…
3,000
2
Sue
…
…
…
500
3
Ann
…
…
…
1,700
4
Jim
…
…
…
1,500
5
Liz
…
…
…
0
6
Dave
…
…
…
9,000
7
Sue
…
…
…
1,010
8
Bob
…
…
…
50
9
Jim
…
…
…
1,300
Extent
Bal Due
page
page
page
1
Bob
…
…
…
3,000
2
Sue
…
…
…
500
3
Ann
…
…
…
1,700
4
Jim
…
…
…
1,500
5
Liz
…
…
…
0
6
Dave
…
…
…
9,000
7
Sue
…
…
…
1,010
8
Bob
…
…
…
50
9
Jim
…
…
…
1,300
Column
Storage
Layout
Customers Table
ID
Name
Address
City
State
Bal Due
1
Bob
…
…
…
3,000
2
Sue
…
…
…
500
3
Ann
…
…
…
1,700
4
Jim
…
…
…
1,500
5
Liz
…
…
…
0
6
Dave
…
…
…
9,000
7
Sue
…
…
…
1,010
8
Bob
…
…
…
9
Jim
…
…
…
Segment =
1 million row chunks
ID
Name
Address
City
State
Bal Due
50
1
Bob
…
…
…
3,000
1,300
2
Sue
…
…
…
500
3
Ann
…
…
…
1,700
4
Jim
…
…
…
1,500
5
Liz
…
…
…
0
6
Dave
…
…
…
9,000
7
Sue
…
…
…
1,010
8
Bob
…
…
…
50
9
Jim
…
…
…
1,300
Run Length Encoding (RLE)
Quarter
ProdID
Price
Quarter
Start
Count
Price
Q1
1
100
Q1
1
310
100
Q1
1
120
Q2
311
290
120
Q1
1
315
…
…
…
315
Q1
1
100
Q1
1
315
Q1
2
198
…
2
450
Q2
2
320
Q2
…
320
Q2
1
150
Q2
1
256
256
Q2
1
450
450
Q2
1
192
Q2
1
184
Q2
2
310
Q2
2
251
251
…
2
266
266
100
ProdID
Start
Count
1
1
5
2
6
3
…
…
…
1
51
5
2
56
3
RLE Compression applied only
when size of compressed data
is smaller than original
315
198
450
320
320
150
192
184
310
Dictionary Encoding
Quarter
Q.ID
Q.ID
Quarter
Q1
1
0
Q1
Q1
1
1
Q2
Q1
1
2
Q3
Q1
1
3
Q4
Q2
2
Q2
2
Q.ID
Start
Count
…
…
1
1
4
2
5
10
Q2
DISTINCT
2
R.L.E.
Only 4 values.
2 bits are enough to
represent it
xVelocity Store
Q3
3
3
11
4
Q3
3
4
15
15
Q3
3
Q3
3
Q4
4
Q4
4
Q4
4
Q4
4
…
…
CPU Architecture
“Batch” Mode
• Modern CPUs have many Cores
• Cache Hierarchies: RAM L3,L2,L1
– Small L1 and L2 per core; L3 shared by socket
– L1 faster than L2, L2 faster than L3
– CPUs stall when waiting for caches to load
• Batch Mode sizes instructions & data to fit into L2/L1 cache !
Parallel Data Warehouse
Overview
Data Warehouse appliances
A prepackaged or preconfigured balanced set of
hardware (servers, memory,
storage and I/O channels),
software (operating system,
DBMS and management
software), service and
support, sold as a unit with
built-in redundancy for high
availability positioned as a
platform for data
warehousing.
Control Node
Failover Protection:
•Redundant Control Node
•Redundant Compute Node
•Cluster Failover
•Redundante Array of
Inexpensive Databases 
Spare Node
Parallel Data Warehouse Appliance - Hardware
Architecture
Corporate Network
Private Network
Storage Nodes
Database Servers
Control Nodes
SQL
Active / Passive
SQL
Client Drivers
SQL
SQL
Management Servers
Landing Zone
Dual Infiniband
Data Center
Monitoring
SQL
SQL
SQL
Dual Fiber Channel
SQL
SQL
ETL Load Interface
SQL
Backup Node
SQL
Corporate Backup
Solution
Spare Database Server
PDW Data Example
PDW Compute Nodes
Time Dim
Product Dim
Date Dim ID
Calendar Year
Calendar Qtr
Calendar Mo
Calendar Day
Prod Dim ID
Prod Category
Prod Sub Cat
Prod Desc
SQL
Sales
Facts
Date Dim ID
Store Dim ID
Prod Dim ID
Mktg Camp Id
Qty Sold
Dollars Sold
SQL
SQL
Mktg
Campaign Dim
Store Dim
Store Dim ID
Store Name
Store Mgr
Store Size
Mktg Camp ID
Camp Name
Camp Mgr
Camp Start
Camp End
SQL
PDW Data Example
Time Dim
Smaller Dimension
Tables are Replicated on
Every Compute Node
Product Dim
Date Dim ID
Calendar Year
Calendar Qtr
Calendar Mo
Calendar Day
Prod Dim ID
Prod Category
Prod Sub Cat
Prod Desc
SQL
Sales
Facts
Date Dim ID
Store Dim ID
Prod Dim ID
Mktg Camp Id
Qty Sold
Dollars Sold
SQL
SQL
Mktg
Campaign Dim
Store Dim
Store Dim ID
Store Name
Store Mgr
Store Size
Mktg Camp ID
Camp Name
Camp Mgr
Camp Start
Camp End
SQL
PDW Data Example
Time Dim
Larger Fact Table is Hash
Distributed Across All
Compute Nodes
Product Dim
Date Dim ID
Calendar Year
Calendar Qtr
Calendar Mo
Calendar Day
Prod Dim ID
Prod Category
Prod Sub Cat
Prod Desc
SQL
Sales
Facts
Date Dim ID
Store Dim ID
Prod Dim ID
Mktg Camp Id
Qty Sold
Dollars Sold
SQL
SQL
Mktg
Campaign Dim
Store Dim
Store Dim ID
Store Name
Store Mgr
Store Size
Mktg Camp ID
Camp Name
Camp Mgr
Camp Start
Camp End
SQL
SMP vs. MPP
SMP: Scale-Up
• Complex – Tune Every IO interface
• Cost – exponential
• Obsolete Hardware
MPP: Scale-Out
Simple – Buy more Processing Nodes
Cost – linear
Keep all hardware investments
SQL Server Parallel Data Warehouse
A quick look at MPP query execution
SQL Server PDW Appliance
steps
are
executed on each compute node
shared nothing
MPP system
connects to
‘the appliance’ like he
would to a ‘normal’
SQL Server
.
generates a
distributed execution
plan
Data Movement
Service
Dealing with Distributions - Shuffling
Compute Node 2
Compute Node 1
Distributed Table
1
Red
5
Red
5
3
Blue
11
Red
8
5
Red
12
Red
12
7
Green
7
Green
7
2
Red
8
Blue
11
4
Blue
10
Blue
10
6
Yellow
12
Yellow
12
Blue
21
Green
7
Red
25
Yellow
12
Overall Architecture
Control Rack
Data Rack (up to 4)
Compute Node 1
Control Node
Client Interface
(JDBC, ODBC,
OLE-DB, ADO.NET)
PDW
Engine
DMS Core
PDW Agent
DMS Manager
Compute Node 2
PDW Agent
DMS Core
PDW Agent
Landing Zone Node
…
ETL
Interface
Bulk Data Loader
PDW Agent
Management Node
Compute Node 10
Active Directory
Legend:
PDW Agent
PDW service
PDW = Parallel Data Warehouse
DMS = Data Movement Service
DMS Core
PDW Agent
(c) 2011 Microsoft. All rights reserved.
Release Themes
How did it work before?
Control Node
•
Problem
– Basic RDBMS functionality, that already exists in SQL Server, was re-built in PDW 
•
Challenge for PDW AU3 release
– Can we leverage SQL Server and focus on MPP related challenges?
SQL Server PDW AU3 Architecture
PDW AU3 Architecture with Shell Appliance and Cost-Based Query Optimizer
SELECT
SELECT
foo
Shell Appliance
(SQL Server)
Every database exists as an empty
‘shell’
• All objects, no user data
Control Node
SQL Server runs a ‘Shell Appliance’
Engine Service
foo
Plan
Steps
Compute Node
(SQL Server)
Plan
Steps
Large parts of basic RDBMS
functionality now provided by the
shell
• Authentication and authorization
• Schema binding
• Metadata catalog
Plan
Steps
DDL executes against both the shell
and the compute nodes
Compute Node
(SQL Server)
Compute Node
(SQL Server)
foo
foo
SQL Server PDW AU3 Architecture
PDW AU3 Architecture with Shell Appliance and Cost-Based Query Optimizer
1. User issues a query
–
SELECT
Shell Appliance
(SQL Server)
MEMO
–
SQL Server performs parsing, binding,
authorization
SQL optimizer generates execution
alternatives
SELECT
Return
Engine Service
3. MEMO containing candidate plans,
histograms, data types is generated
Plan
Steps
Plan
Steps
Plan
Steps
4. Parallel execution plan generated
Control Node
2. Query is sent to the Shell through
sp_showmemo_xml stored procedure
Compute Node
(SQL Server)
Compute Node
(SQL Server)
5. Parallel plan executes on compute
nodes
6. Result returned to the user
Compute Node
(SQL Server)
PDW Cost-Based Optimizer
Optimizer lifecycle…
1. Simplification and space exploration
– Query standardization and simplification (e.g. column reduction, predicates pushdown)
– Logical space exploration (e.g. join re-ordering, local/global aggregation)
– Space expansion (e.g. bushy trees – dealing with intermediate resultsets)
– Physical space exploration
– Serializing MEMO into binary XML (logical plans)
– De-serializing binary XML into PDW Memo
2. Parallel optimization and pruning
– Injecting data move operations (expansion)
– Costing different alternatives
– Pruning and selecting lowest cost distributed plan
3. SQL Generation
– Generating SQL Statements to be executed
PDW Cost-Based Optimizer
… And Cost Model Details
• PDW cost model assumptions:
– Costing only data movement operations (relational operations excluded)
– Sequential step execution (no pipelined and independent parallelism)
• Data movement operations consist of multiple
tasks
• Each task has Fixed and Variable overhead
• Uniform data distribution assumed (no data
skew)
PDW Sales Test Workload
AU2 to AU3
80
70
60
50
AU2
40
AU3
30
20
10
0
1
•
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
5x improvement in terms of total elapsed time out of the box
29
30
31
32
33
34
35
36
37
38
39
Theme: Performance at Scale
Zero data conversions in data movement
DMS CPU Utilization - TPCH
60
50
•
40
Functionality
•
•
Using ODBC instead of ADO.NET for
reading and writing data
Minimizing appliance resource utilization
for data moves
Benefits
•
•
•
•
Better resource, CPU, utilization
6x or more faster move operations
Increased concurrency
Mixed workload (loads + queries)
30
20
10
0
Q1
Q2
Q3
Q4
Q5
Q6
Q7
Q8
Q9
Q10
Q11
Q12
Q13
Q14
Q15
Q16
Q17
Q18
Q19
Q20
Q21
Q22
•
Eliminate CPU utilization spent on data
conversions
Further parallelize operations during data
moves
CPU (%)
Goal
AU2
AU3
Throughput improvement for data movements
Repl Table Load
Shuffle
Replicate
Trim
Broadcast
0%
200%
400%
600%
Theme: SQL Server Compatibility
SQL Server Security and Metadata
Security
•
•
•
•
•
SQL Server security syntax and semantics
Supporting user, roles and logins
Fixed database roles
Allows script re-use
Allows well-known security methods
Metadata
•
•
•
•
•
PDW metadata stored in SQL Server
Existing SQL Server metadata tables/views (e.g. security views)
PDW distribution info as extended properties in SQL Server metadata
Existing means and technology for persisting metadata
Improved 3rd party tool compatibility (BI, ETL)
Theme: SQL Server Compatibility
Support for SQL Server (Native) Client
Goal
•
•
‘Look’ just like a normal SQL Server
Better integration with other BI tools
Functionality
•
•
•
•
Use existing SQL Server drivers to connect
to SQL Server PDW
Implement SQL Server TDS protocol
Named Parameter support
SQLCMD connectivity to PDW
Benefits
•
•
•
•
Use known tools and proven technology
stack
Existing SQL Server ’eco-system’
2x performance improvement for return
operations
5x reduction of connection time
Server: 10.217.165.13, 17001
SQL Server Clients
TDS
(ADO.NET, ODBC,
OLE-DB, JDBC)
SequeLink
SQL PDW Clients
(ODBC, OLE-DB,
ADO.NET)
Server: 10.217.165.13, 17000
Theme: SQL Server Compatibility
Stored Procedure Support (Subset)
Goal
•
Support common scenarios of code
encapsulation and reuse in Reporting and ETL
Functionality
•
•
•
System and user-defined stored procedures
Invocation using RPC or EXECUTE
Control flow logic, input parameters
Benefits
•
•
•
•
Enables common logic re-use
Big impact for Reporting Services scenarios
Allows porting existing scripts
Increases compatibility with SQL Server
Theme: SQL Server Compatibility
Collations
Goal
•
Support local and international data
Functionality
•
•
•
•
Fixed server level collation
User-defined column level collation
Supporting all Windows collations
Allow COLLATE clauses in Queries and DML
Benefits
•
•
•
Store all the data in PDW w/ additional
querying flexibility
Existing T-SQL DDL and Query scripts
SQL Server alignment and functionality
Theme: Improved Integration
SQL Server PDW Connectors
Connector for Hadoop
•
•
•
•
•
Bi-directional (import/export) interface between MSFT Hadoop and PDW
Delimited file support
Adapter uses existing PDW tools (bulk loader, dwsql)
Low cost solution that handles all the data: structured and unstructured
Additional agility, flexibility and choice
Connector for Informatica
•
•
Connector providing PDW source and target (mappings, transformations)
Informatica uses PDW bulk loader for fast loads
•
Leverage existing toolset and knowledge
Connector for Business Objects
Agenda
•
•
•
•
Trends in the DW space
How does SQL Server PDW fit in?
SQL Server PDW AU3 – What’s new?
Building BI Solutions with SQL Server PDW
–
–
–
–
Customer Successes
Using SQL Server PDW with Microsoft BI solutions
Using SQL Server PDW with third party BI solutions
BI solutions leveraging Hadoop integration
• What’s coming next in SQL Server PDW?
PDW Retail POS Workload
Original Customer SMP solution vs. PDW AU3 (with cost-based query optimizer)
1600
1400
1200
1000
Old SMP
800
POS ODS AU3
600
400
200
0
Q1
Q2
Q3
Q4
Q5
Q6
Q7
How are customers using PDW & BI ?
Data Volume
•
•
80 TB data warehouse analyzing data from
exchanges
Existing system based on SQL SMP farm
–
2 different clusters of 6 servers each
Portal
Requirement
•
•
•
Linear scalability with additional hardware
Support hourly loads with SSIS – 300GB/day
BI Integration: SSRS, SSAS and PowerPivot
Reports
ETL
Dashboards
PDW
AU3 Feedback
•
•
•
•
SP and increased T-SQL support was great
Migrating SMP SSRS to PDW was painless
142x for scan heavy queries & no summary tables
Enabled queries that do not run on existing system
Operational
DB’s
Scorecards
Fast
parallel
Infiniband
Aggregation abilities avoids ETL overhead in existing systems
No need
indexes
indexed/materialized views
Infiniban
d
GBit
link
SSAS with SQL Server PDW
Understanding the differences compared to ‘SMP world’
foreign key constraints
data design
retrieval planning
MOLAP & ROLAP
Specific to the nature of large data
limits
include required data
New Challenges for Business Analytics
•
•
•
Huge amount of data born ‘unstructured’
Increasing demand for (near) real-time
business analytics
Pre-filtering of important from less relevant
raw data required
HADOOP
Applications
•
•
•
Sensor networks & RFID
Social networks & Mobile Apps
Biological & Genomics
Sensor/ Blogs,
RFID Data Docs
Web
Data
Hadoop as a Platform Solution
In the context of ETL , BI , and DW
•
Platform to accelerate ETL processes
(not competing with current ETL
software tools!)
•
Flexible and fast development of ‘handwritten’ refining requests of raw data
Fast ETL
processing
Cost-Optimal
storage
•
Active & cost effective data archive to let
(historical) data ‘live forever’
•
Co-existence with a relational DW
(not completely replacing it !)
HADOOP
Fast
Refinery
Active Archi
Importing HDFS data into PDW for advanced BI
Application
Programmers
DBMS
Admin
Power
BI Users
SQOOP
HADOOP
Sensor/ Blogs, Web
RFID Data Docs Data
SQL Server PDW
Interactive BI/Data
Visualization
Hadoop - PWD Integration via SQOOP (export)
SQOOP export with
source (HDFS path) &
target (PDW DB & table)
2.
Read HDFS
data via mappers
Copies incoming data
on Landing Zone
3.
1.
PDW
Hadoop
Connector
FTP
Server
5.
…
4.
Telnet
Server
…
HDFS
Invokes
‘DWLoader’
Compute
Node 1
Landing Zone
PDWconfiguration file
Linux/ Window
Hadoo s/PDW
ControlCompute
Node Nodes
Compute
Node 8
SQL Server PDW Roadmap
What is coming next?
CALENDAR YEAR 2011
Q1
Shipped
Appliance Update 1
• Improved node
manageability
• Better performance and
reduced overhead
• OEM requests
Q3
Q2
Shippe
d
CALENDAR YEAR 2012
Q4
Appliance Update 2
• Programmability
• Batches
• Control flow
• Variables
• Temp tables
• QDR infiniband switch
• Onboard Dell
Q1
Shipped
Q2
Q3
Q4
V-Next
Appliance Update 3
• Cost based optimizer
• Native SQL Server drivers,
including JDBC
• Collations
• More expressive query language
• Data Movement Services
performance
• SCOM pack
• Stored procedures (subset)
• Half-rack
• 3rd party integration
(Informatica, MicroStrategy,
Business Objects, HADOOP)
•
•
•
•
•
•
•
•
•
Columnar store index
Stored procedures
Integrated Authentication
PowerView integration
Workload management
LZ/BU redundancy
Windows 8
SQL Server 2012
Hardware refresh

similar documents