Robin SchumacherVP Products
Calpont
Calpont’s InfiniDB
Who the Heck are
We…?
InfiniDB by Calpont is…
A column-oriented database engine A great option for analytics, data warehousing, and read-intensive apps A scale up and out design that future proofs your systems
InfiniDB scales up…It will address all the CPU’s/cores on your machine.
SSB Query (@100 scale)
InfiniDB 1 Core(elapsed time in
seconds)
InfiniDB 8 cores
(elapsed time in seconds)
Overall Percent
Reduction with
additional cores
Q2.1 210.21 44.65 79%
Q2.2 151.20 19.70 87%
Q2.3 121.33 15.94 87%
Q3.1 316.79 55.04 83%
Q3.2 164.12 22.14 87%InfiniDB runs great on a single machine as well as on multiple boxes.
InfiniDB scales out…It does MPP across multiple nodes, with auto-failover.
SSB Query@1000
1 PM(elapsed time in seconds)
2 PM(elapsed time in seconds)
4 PM(elapsed time in seconds)
8 PM(elapsed time in seconds)
Overall Percent Reduction from
1 – 8 PM’s
Q2.1 531.34 261.35 129.90 68.21 87%
Q2.2 430.25 214.87 106.37 56.41 87%
Q2.3 386.66 192.03 96.03 51.36 87%
Q3.1 848.79 425.25 316.50 134.21 84%
Q3.2 597.97 297.46 148.49 77.74 87%
PM = Performance Module, which handles I/O operations for InfiniDB
InfiniDB has…Plenty of features to support demanding environments• MySQL front end
• Column-oriented• Multi-Terabyte database capable• High-Speed bulk loader w/ no blocking queries while
loading• Crash-recovery• Transaction support (ACID compliant)• INSERT/UPDATE/DELETE (DML) support• No indexing necessary• Automatic partitioning of data• Hash joins• MVCC support – snapshot read (readers don’t block
writers)• Alter Table with online add column capability • High concurrency supported• Logical data compression• Multi-Node, MPP scale out capable w/ failover• GPL and Commercial EditionsInfiniDB is designed for fast load and query performance
How Fast…?
InfiniDB vs. “Leading” Row DatabaseInfiniDB takes up 22% less space InfiniDB loaded data 22% faster
InfiniDB total query times were 65% less InfiniDB average query times were 59% less
Notice not only are
the queries faster, but also more
predictable
* Tests run on standalone machine: 16 CPU, 16GB RAM, CentOS 5.4 with 2TB of raw data
Percona’s Test of Open Source Column Databases
610 GB of raw data; 8 Core Machinehttp://www.mysqlperformanceblog.com/2010/01/07/star-schema-bechmark-infobright-infinidb-and-luciddb/
(lower is better)
Internal MPP Stress Test using SSB Benchmark
Six Performance Modules against data size of Percona tests and then 10x the size. Shows predictable performance, MPP linear speed gains over standalone implmentation, and capabilities
that other open source databases don’t have.http://www.infinidb.org/infinidb-blog/mysql-parallel-query-processing-of-60-billion-rows-via-infinidb.html
Predictable Performance with More Data
What else should I know…?
Connectors and More…From a dev perspective, use InfiniDB as you would standard MySQL
All connectors, tools, dev IDE’s, etc., should work fine with InfiniDB
Use InfiniDB & Other Engines…
Other MySQL engines can be used with InfiniDB in the same database/instance
Use InfiniDB & Other Engines…
Other engines can also be used with InfiniDB in an MPP fashion
Our Commitment…
1.0 now available for all
Release early/often with monthly maintenance updates for Community Go where the Community and Customers send us
What is coming next…?
Coming Attractions for InfiniDB…
July 2010: InfiniDB 1.1High-speed subqueries, UTF8, Windows port, and more.
December 2010: InfiniDB 1.5-2.0Shared nothing support, physical compression, and more.
2011: InfiniDB 2.x-3.xYou tell us…!
How can I get it…?
• Download InfiniDB Community Edition• Download/View Free InfiniDB documentation• Read InfiniDB technical white papers and articles• Read InfiniDB intro articles on MySQL dev zone• Visit InfiniDB online forums• Trial the InfiniDB Enterprise Edition:
http://www.calpont.comwww.infinidb.org