Lenovo Database Configuration for Microsoft SQL Server 2016 – 22TB

Lenovo Database Configuration for
Microsoft SQL Server 2016 – 22TB
Data Warehouse Fast Track Solution
Data Warehouse problem and a solution
The rapid growth of technology means that the amount of available data and the
ability to collect that data increased to a level unthinkable as little as five years ago.
As the volume and velocity of data increased, however, extracting meaningful insight
in a timely manner became more complex. Therefore, opportunities are being
missed and effort is being wasted. To compete, businesses in the 21st century are
demanding the tools to derive true value from their data.
This Microsoft Data Warehouse Fast Track (DWFT) configuration for SQL Server
2016 improves time-to-value for data warehousing needs with a new scalable
architecture. This solution in the Lenovo portfolio uses the high performance System
x3650 M5 server combined with NVMe Flash Adapter Storage to solve SQL
database warehouse needs up to 22 TB in size.
The Microsoft Data Warehouse Fast Track program makes it easy to reduce costs,
save time, and reduce risk with reliable, pretested hardware and best practices for
data warehousing. This solution features the following highlights:
Reduce time to value with pretested hardware configurations.
Reduce hardware testing and reduce tuning immediately.
Reduce total cost of ownership through better price and performance, rapid
deployment, and advanced hardware.
Optimize performance with pretested System x3650 M5 hardware configurations.
Consolidate storage and match IT investment-to-information-value with NVMe
Flash adapters.
The configuration listed in this document has a Fast Track RowStore rating of I/O:
8,305 MBps and ColumnStore throughput: of 1,851 Queries/Hr/TB.
Enterprise data warehouse with faster time-to-value
DWFT for SQL Server 2016 for Lenovo solution offerings are methodically tested
and tuned to save you months of configuration, setup, testing, and tuning. With these
offerings from Lenovo, you can now complete the following tasks:
Buy all the hardware that you need from only one vendor including servers,
storage, and networking.
Build, tune, and deploy with confidence by using established data warehouse
best practices.
Select from different levels of performance, scalability, and price to suit your
business needs.
Choose from 4 to up to 120 Intel Xeon processor cores.
Run targeted query workloads that are patterned for large sequential data sets
rather than small random transactions.
Eliminate bottlenecks with optimized rapid data reads and query aggregations
 Standard 22 TB SQL
DWFT solution from
 Balanced and optimized
 High performance NVMe
storage solution
 Certified by Microsoft
 Reduced time to value
Lenovo Database Configuration for Microsoft SQL Server 2016 – 22TB
Microsoft SQL Server 2016
Microsoft® SQL Server 2016 has made significant improvements in data
warehousing technologies and performance, including column-store features as
well as many other improvements. Column-store indices offer great advantages
over traditional row stores for analytics and data warehousing queries. They are
ideally suited for the star schemas, and tables with billions of rows which are
commonly seen. Among their advantages for analytics are:
Up to 10X compression in data size - Data warehouses are very large by
nature, and the compression offered by column store index technologies offers
both space and cost savings, but also significantly increased performance,
due to the dramatically reduced IO requirements given by the compression,
coupled by the ability to only scan the specific columns required by each
query. Compression also reduces the amount of memory required to hold a
given number of rows from the source data warehouse.
Additional Indices - SQL Server 2016 adds the capability to add additional
(B-Tree) indices to column store-based tables, which enables efficient singlerow lookup.
In addition to these architectural features, query processing in column-store
indices is further optimized in the following ways:
Operator Pushdown - Pushdown refers to moving both filter and aggregation
query operations closer to the data, so that many of the filters and calculations
can be done in the scan operators, dramatically reducing the volume of data
which needs to be handled further on in query processing.
Batch Mode Processing - SQL Server 2016 includes enhancements in
batch-mode processing which processes many rows at a time rather than
serially doing calculations on each individual row. These batch operations are
further optimized by leveraging Single Instruction Multiple Data (SIMD) vector
processing CPU instructions in the Intel® architectures.
Configuration tested and certified
This configuration features the following main components:
Server: Lenovo System x3650 M5
Processor: Two E5-2699 v4 22C 2.2 GHz
Memory: 512 GB of DDR4 memory
Storage: Four 2.0TB NVMe Flash Adapters for data and tempdb
OS Storage: Two 300 GB SAS HDDs for the operating system (RAID 1)
Logging: Two 480 GB SSDs for log (RAID 1)
 Microsoft Windows Server 2016 Standard Edition
 Microsoft SQL Server 2016 Enterprise Edition
Lenovo Database Configuration for Microsoft SQL Server 2016 – 22TB
Powered by System x3650 M5 and 2.0TB NVMe
Flash Adapters
Lenovo System servers, such as the System x3650 M5 server, feature the latest
Intel Xeon E5-2600 v4 series processors. With more cores and more memory, the
new M5 family is fast. The greatly increased processing power is provided by the
latest Intel Xeon E5-2600 v4 processors. Lenovo System x3650 M5 servers include
the following features:
Two times the memory capacity of previous generation processors, with 24
DIMM sockets in the x3650 M5.
Support for 64 GB TruDDR4 Memory LRDIMMs, up to 1.5 TB of memory in the
x3650 M5.
New storage technologies, such as the Intel P3700 2.0TB NVMe Enterprise
Performance Flash Adapter for Lenovo System servers, which closely align the
performance of storage with the power of the processors.
o As an alternate to the Intel P3700 Flash Adapter, a quantity of four Lenovo
PX04PMC 1.92TB NVMe Mainstream Flash Adapters may be employed in
this configuration.
DWFT for SQL Server 2016 features the System x3650 M5 with 2.0TB NVMe Flash
Adapters that improve productivity through data consolidation, availability
performance, and scalability. These solid-state adapters simplify DWFT storage
configuration and maintenance versus the use of a SAN, which has more parts to
maintain and manage. The reference configuration is a 2-socket system that uses
the DWFT V4 methodology.
Best practices for Data Warehouse Fast Track
For a balanced and optimized Data Warehouse configuration:
Configure UEFI settings to set Memory mode to Independent.
Configure UEFI settings to set Operating mode to Maximum performance.
Configure high availability for the OS with 2-disk Raid-1.
Configure high availability for the log drive with 2-disk Raid-1 or Raid-10 with
more disks based on performance needs.
Data files and tempdb can be on Raid 0 drives. Spread data and tempdb files
evenly across all data drives for even performance.
Configure more than one tempdb files; at least one file per data drive.
Enable lock pages in memory option using Windows Group policy tool to prevent
paging of data.
If the server is dedicated to data warehousing,
 Set processor affinity for SQL Server to use all the processors in the system.
 Set SQL Server Maximum Server Memory to 90% of the total memory
available on the server.
 Add –E and optionally –T834 to SQL Server Startup parameters.
Lenovo System x3650 M5 High
Performance Rack Server, Intel P3700
NVMe PCIe adapter and Lenovo
NVMe Mainstream Flash Adapter
Lenovo Database Configuration for Microsoft SQL Server 2016 – 22TB
Reference Architecture Certification
Lenovo Database Configuration for Microsoft SQL Server 2016 – 22TB
Lenovo Database Configuration for Microsoft SQL Server 2016 – 22TB
High Performance at low cost with Lenovo
x3650 M5 servers and NVMe PCIe
Why Lenovo System servers for Microsoft SQL
Lenovo offers a wide range of servers and options. The Lenovo reference
configurations for DWFT for SQL Server bring together the right mix of technology
and software. The configurations integrate the latest powerful Lenovo System rack
and enterprise servers, robust Lenovo Storage options, and the data warehouse
capabilities of SQL Server 2016 Enterprise Edition.
Why Lenovo
Lenovo is a leading provider of x86 servers for the data center. Featuring rack,
tower, blade, dense and converged systems, the Lenovo server portfolio provides
excellent performance, reliability and security. Lenovo also offers a full range of
networking, storage, software, solutions, and comprehensive services supporting
business needs throughout the IT lifecycle. With options for planning, deployment,
and support, Lenovo offers expertise and services needed to deliver better servicelevel agreements and generate greater end-user satisfaction.
For More Information
To learn more about the Lenovo Database Configuration for Microsoft SQL Server
2016 – 22 TB solution, contact your Lenovo Business Partner or visit:
For more information on the Lenovo NVMe Mainstream Flash Adapters, visit:
© 2017 Lenovo. All rights reserved.
Availability: Offers, prices, specifications and availability may change without notice. Lenovo is not responsible for photographic or typographical errors. Warranty: For a
copy of applicable warranties, write to: Lenovo Warranty Information, 1009 Think Place, Morrisville, NC, 27560, Lenovo makes no representation or warranty regarding thirdparty products or services. Trademarks: Lenovo, the Lenovo logo, System x, ThinkServer are trademarks or registered trademarks of Lenovo. Microsoft and Windows are
registered trademarks of Microsoft Corporation. Intel, the Intel logo, Xeon and Xeon Inside are registered trademarks of Intel Corporation in the U.S. and other countries.
Other company, product, and service names may be trademarks or service marks of others. CRN: DBSSQLM2262