High performance analytics
Thousands of industry leaders
rely on MicroStrategy.
The top six reasons why
MicroStrategy is a performance leader.
Terabyte scale
In-memory BI
Fastest
dashboards
with up to 60% reduction
in bandwidth
Highly
optimized
High
performance
on federated sources
High availability
and self-optimizing platform
MicroStrategy
PRIME
Raising the bar
in high performance analytics.
1.
Analytics
complexity
MicroStrategy
can do all three:
2.
Big data
3.
High user
scale
Users expect much higher performance from applications today. 57% of users
will abandon an application if they have to wait more than three seconds for the page to load. And this percentage increases when you factor in mobile.
250,000+ users
supported with an average
sub 2-second response.
10x higher
median customer data
volume than the average of the
next ten vendors.*
Linear performance
that increases as number of cores
increase for both virtualized and
non-virtualized environments.
The volume, variety, and velocity of data is exploding. 90% of the world’s data has
been generated over the past two years.
Analytic applications are becoming much more complex. Most successful
analytic applications have 150+ KPIs.
High performance is no longer a luxury
but a necessity.
Analytics apps are engaging more users in the enterprise. Increasing from 10%
to 80% with the addition of self-service and mobile.
Performance Benchmark:
Enterprise Analytics.
The key to making better business decisions is the ability to deliver the right information at the right time to the right user. MicroStrategy’s comprehensive analytics platform delivers high performance self-service and enterprise BI on a single platform, on web or mobile devices, and on-premises or in the cloud.
Overall results:
Using an 8-CPU configuration on v9.4 • Sub 2-second response
• 1.9x greater capacity than v9.0 • Test supported 50,710 active users
(253,550 actual users)
Performance Benchmark:
Enterprise Analytics.
Performance tests were conducted using a standard benchmark application that reproduced the usage pattern of a typical enterprise business intelligence application that includes reporting, dashboards, and visual data discovery with simulated user populations of up
to 255,000 users. The test application consisted of a range of 1,000 to over 60,000 active users accessing in-memory datasets derived from a 1TB database that contained 7B transaction-level records.
ACTIVE USERS**
Power rating comparison for:
Configuration: 8 CPUs, Linux 64, Enterprise Application Mix
0 9,000 18,000 27,000 36,000 45,000 54,000 63,000 0 1 2 3 4 5 0 50 100 150 200 250 300 350
QUERY LOAD (kilocycles) 1.9x greater capacity
Power Rating = 284 Kilocycles Active users supported = 50,710
Version 9.0 Version 9.4
RESPONSE TIME
(sec)
For more details on the test set-up
and more technical information, visit
microstrategy.com/performance
Performance Benchmark:
Self-service visual data discovery.
Self-service analytics allows business users to visually explore data and spot patterns and outliers. With MicroStrategy’s high performance self-service tools, users can quickly mash up data from multiple sources and develop insights. Creating brilliant and intuitive visualizations is fast, easy, and powerful.
Visual data discovery:
Using an 8-CPU configuration on v9.4 • Sub 3-second response
• 2.1x greater capacity than v9.3 • Test supported 6,370 active users
(31,850 actual users)
Performance Benchmark:
Self-service visual data discovery.
Performance tests were conducted using a standard benchmark application that reproduced the usage pattern of a typical visual data discovery-based
analytics application. The application includes analysis creation, execution, manipulation, storage, and deletion of data based on visual interactive analysis.
Power rating comparison for:
Configuration: 8 CPUs, Linux 64, Visual data discovery
0 1 2 3 4
5 Power Rating = 118 KilocyclesActive users supported = 6,370 2.1x greater capacity
Version 9.4 Version 9.3 6
0 20 40 60 80 100 120
QUERY LOAD (kilocycles)
140
ACTIVE USERS**
0 1,080 2,160 3,240 4,320 5,400 6,480 7,560
RESPONSE TIME
(sec)
The test application consists of a range of 1,000 to over 7,000 active users accessing in-memory datasets.
For more details on the test set-up
and more technical information, visit
microstrategy.com/performance
Performance Benchmark:
Dashboards and information-driven apps.
Dashboards and information-driven apps allow organizations to embed analytics at the point of decision making. MicroStrategy’s powerful dashboard engine allows customers to create applications that tightly integrate into existing workflows, and are personalized to thousands of users. Delivering high information density through a guided business workflow, high performance information-driven apps from MicroStrategy are what’s next for dashboards.
Dashboard results:
Using an 8-CPU configuration on v9.4 • Sub 3-second response
• 1.4x greater capacity than v9.3 • Test supported 1,770 active users
(8,850 actual users)
Performance Benchmark:
Dashboards and information-driven apps.
Performance tests were conducted using a typical business intelligence application that focuses on dashboard execution and manipulations in web.
The test application consists of a range of 300 to over 2,000 active users accessing in-memory datasets.
ACTIVE USERS**
Power rating comparison for:
Configuration: 8 CPUs,Win 64, Dashboard applications
0 270 540 810 1,080 1,350 1,620 1,890 0 1 2 3 4 5 RESPONSE TIME (sec) 0 5 10 15 20 25 30 35
QUERY LOAD (kilocycles)
1.4x greater capacity Version 9.3 Version 9.4 6 7 8 40 2,160
Power Rating = 33 Kilocycles Active users supported = 1770
For more details on the test set-up
and more technical information, visit
microstrategy.com/performance
Performance Benchmark:
OLAP analytics.
Users often have to manipulate data to gather insights. Actions such as sorting, pivoting, drilling through, and filtering are commonly referred to as OLAP analytics. Only MicroStrategy can deliver high performance OLAP analytics against multi-terabyte datasets through a unique combination of relational OLAP (ROLAP) and in-memory technology.
OLAP analytics results:
Using an 16-CPU configuration of MAE v9.4 • Sub 2-second response
• Test supported 10,600 active users (53,000 actual users)
• Peak power rating of 59 Kilocycles*
Performance Benchmark:
OLAP analytics.
Performance tests were conducted using a typical BI application that simulated users performing intensive OLAP manipulations and interactions. These included (but were not limited to)
pivots, drills, filters, and other interactions using complex metrics. The test application consisted of a range of 1,000 to over 12,000 active users accessing in-memory datasets.
ACTIVE USERS**
Power rating comparison for:
Configuration: 8 CPUs, Linux 64, OLAP application
0 1,800 3,600 5,400 7,200 9,000 10,800 12,600 0 1 2 3 4 5 RESPONSE TIME (sec) 0 10 20 30 40 50 60 70
QUERY LOAD (kilocycles)
Power Rating = 59 Kilocycles Active users supported = 10,600
Version 9.4
For more details on the test set-up
and more technical information, visit
microstrategy.com/performance
Performance Benchmark:
Mobile BI - iPad.
Users now expect the same high performance from analytics apps on their mobile device thanks to their experience with products from Apple and Google. Business users need to be able to explore, analyze, and act on their data wherever they are, not just when they are at their desk, or when they have a Wi-Fi connection. MicroStrategy’s #1-rated mobile BI platform delivers the high performance that users have come to expect on the iPad.
iPad results:
Using an 8-CPU configuration on v9.4 • Sub 2-second response
• 1.4x greater capacity than v9.3 • Test supported 29,140 active users
(145,700 actual users)
• Peak power rating of 540 Kilocycles*
iPad 12:30 PM 100%
Performance Benchmark:
Mobile BI - iPad.
Performance tests were conducted on a mobile business intelligence application running on an iPad. The applicatioin includes various mobile dashboard executions. The test application consisted of a range of 2,000 to over 32,000 active users accessing in-memory datasets.
For more details on the test set-up
and more technical information, visit
microstrategy.com/performance
ACTIVE USERS**Power rating comparison for:
Configuration: 8 CPUs, Win 64, iPad application
0 5,400 10,800 16,200 21,600 27,000 32,400 37,800 0 1 2 3 4 5 RESPONSE TIME (sec) Version 9.4 6 0 100 200 300 400 500 600 700
QUERY LOAD (kilocycles)
Power Rating = 540 Kilocycles Active users supported = 29,140
1.4x greater capacity
Performance Benchmark:
Mobile BI - iPhone.
A mobile workforce means users have access to their data and analytics 24x7. MicroStrategy exhibits an average sub 2-second response time for BI apps on an iPhone even under heavy load. As a result, great performance and analytics that users have come to expect on the iPhone from consumer apps is now available for analytics apps as well.
Overall results:
Using an 8-CPU configuration of v9.4 • Sub 2-second response
• 1.3x greater capacity than v9.3 • Test supported 27,090 active users
(135,450 actual users)
• Peak power rating of 510 Kilocycles*
12:30 PM 100% 12:30 PM 100%
Performance Benchmark:
Mobile BI - iPhone.
Performance tests were conducted on a mobile business intelligence application running on an iPhone. The application includes various mobile dashboard
execution scenarios. The test application consisted of a range of 2,000 to over 30,000 active users accessing in-memory datasets.
ACTIVE USERS**
Power rating comparison for:
Configuration: 8 CPUs, Win 64, iPhone application
0 5,400 10,800 16,200 21,600 27,000 32,400 0 1 2 3 4 5 RESPONSE TIME (sec) 0 100 200 300 400 500 600
QUERY LOAD (kilocycles)
Power Rating = 510 Kilocycles Active users supported = 27,090
1.3x greater capacity
Version 9.4 Version 9.3
For more details on the test set-up
and more technical information, visit
microstrategy.com/performance
Performance Benchmark:
Performance impact of increasing CPU cores and virtualization.
Analytic applications rapidly growas organizations embrace business intelligence as a core element of success. A well-architected analytics solution should scale linearly and predictably to allow enterprises to plan and account for growth. In addition, analytic platforms should run well on both physical servers and virtual servers for maximum flexibility. MicroStrategy’s analytics platform uniquely satisfies both these requirements.
Results:
• Performance increases linearly as the number of cores increase for both virtualized and non-virtualized environments.
• Running a BI application on a Linux operating system within a virtual machine (VM) imposes a performance penalty that varies according to the number of CPUs in the server with 9% reduction at 1 CPU, 11% reduction in capacity at 4 CPUs and 13% reduction at 8 CPUs.
• Running in a VM environment did not affect the average response time performance of the BI application. Both systems delivered the same sub 3.5- second average response times.
Performance Benchmark:
Performance impact of increasing CPU cores and virtualization.
Performance tests were conducted using a typical business intelligence application to measure the impact of increasing the number of CPU cores as well as the impact of using virtualization technology.
The test application consisted of a range of 300 to over 2,000 active users accessing in-memory datasets.
Power rating comparison for:
Configuration: Dashboard App, MicroStrategy v9.3
60 50 40 30 20 10 11% Linux on VM Linux 0 0 1 2 3 4 5 6
Number of CPU Cores
7 PO WER R ATING* (K iloc ycles) A C TIVE USERS* 8 9 3,240 2,700 2,160 1,620 1,080 540 13%
For more details on the test set-up
and more technical information, visit
microstrategy.com/performance
Introducing MicroStrategy PRIME:
Game changing in-memory technology.
MicroStrategy PRIME breaks new groundby tightly coupling a state-of-the art visualization and dashboarding engine with an innovative massively parallel in-memory data store. This architecture allows companies to rapidly build and deploy powerful information-driven apps that deliver analytics to hundreds of thousands of users in a fraction of the time and cost of other approaches.
Married College Urban
41% 77% 80%
Wisdom Summary
CloseClose Filter
Filter results: 15,131,760 fans in Wisdom
Interests: Demographics:
Location & Language: Average Fan Age – 32
Under 17 17–28 29–35 36–45 Over 45 4% 45% 20% 17% 14%
48% of fans are women
52% of fans are men WISDOM OVERVIEW 15,131,760FANS 1,475,778,873 PAGE LIKES 141,601 CITIES 98
AVG PAGE LIKES PER FAN
332
AVG # OF FRIENDS PER FAN
32 AVG AGE Overall 1. Facebook 2. Barack Obama 3. Family Guy 4. YouTube 5. Michael Jackson Movies 1. Megan Fox 2. Will Smith 3. The Hangover 4. Harry Potter 5. Vin Diesel Sports 1. Cristiano Ronaldo 2. Michael Jordan 3. Leo Messi 4. FC Barcelona 5. NBA Music 1. Michael Jackson 2. Eminem 3. Lady Gaga 4. Bob Marley 5. The Beatles Companies/Products 1. Facebook 2. YouTube 3. Starbucks 4. Coca-Cola 5. Disney
Facebook runs PRIME:
• 300+ PB of Hadoop source data • 30+ TB analyzed in PRIME • 220+ nodes
• 175 billion rows
Achieve unprecedented performance with MicroStrategy PRIME
0 1 2 3 4 5 A VER A GE RESPONSE TIME (sec) 0 2 4 6 8 10 12POWER RATING (kilocycles)
Well-known database appliance Well-known in-memory database 6 7 8 9 14 PRIME 7x more users 3x faster
• High user interactivity • 200GB data set with 50+
dimensions
• Complex analytical dashboard • Equivalent hardware
Big Data
Performance.
MicroStrategy customers analyze 10 times more data than the industry average. In addition to terabytes of structured data, they also integrate personal, cloud, and unstructured data into their analytic applications. MicroStrategy’s unique multi-Median Customer Data Volumes*
MicroStrategy
Competitor Average
0 100 200 300 400 500 600 GBs
source capabilities allow organizations to present a single consolidated view of all this information to users without physically integrating and moving all the data. With our patented ROLAP architecture, queries are optimized for each individual data
source, and industry leading in-memory technology allows for sub-second responses on large federated data sets.
MapReduce Databases and NoSQL Hortonworks Cloudera Aster Data MapR Amazon Elastic Map Reduce IBM MongoDB Intel Distribution Pivotal HD Columnar Databases Amazon Redshift Infobright Exasol Kognitio Vertica Sand Google Bigquery ParAccel Calpont Vectorwise Sybase IQ Data Warehouse Appliances TeraData Netezza Oracle Exadata Microsoft SQL Server Parallel Data Warehouse SAP Hana Relational Databases IBM DB2 MySQL IMB Informix Oracle Microsoft SQL Server Sybase PostgreSQL Multidimensional Databases SAP Hyperion Microsoft SQL Server Analytics Services Cognos TM1 SaaS-Based App Data Salesforce Connection Cloud Netsuite Zendesk Zuora Intacct Facebook Google Eloqua User/ Departmental Data Microsoft Excel Text files Microsoft Access .CSV files
Insightful, beautiful,
and effortless:
Self-service visual
data discovery
The world’s most comprehensive
analytics platform. Period.
Trusted analytics
you won’t outgrow:
BI and information-
driven apps
Flexible, agile, and
low risk:
On-premises and in
No data left behind:
Personal data, corporate
data and Big Data
The world’s most comprehensive
analytics platform. Period.
Product Capabilities — Q4 2013
BI Platforms
Ease of Use BusinessQuery DiscoveryVisual
SAP Business Objects
Dashboards Interactive Reports Mobile BI Information Delivery ViewerOLAP
IBM Cognos Oracle BI EE QlikView Tableau Specialt y Production
Reporting PlatformOLAP IntegrationOffice Administration Architecture Cloud BIPublic
MicroStrategy
Legend: Excellent Good Limited Minimal/None Does not compete in this segment © BI Scorecard 2013
BI Scorecard evaluations are the only independent reviews of analytics solutions based on hands-on testing.
Don’t take our word for it. The experts agree:
MicroStrategy has one of the most integrated BI platforms and has continuously expanded its BI capabilities through innovation