making mysql 5,888.6x faster! yep, you read that right!
TRANSCRIPT
Titus Blair - Evangelist at BigObject
@TITUSBLAIR
THE 1,000x EXPERIMENT We wanted to demonstrate the sheer power of BigObject and Pre-Cache technologies but needed a compelling benchmark. We knew that w e c o u l d e a s i l y i n c r e a s e d a t a b a s e performance by a factor of 100x but could we go higher? Could we achieve 1,000x? We love a challenge so we got to work!
THE SYSTEM
• Windows PC
• Intel Core i5–4200H 2.80GHz
• Dual Core up to 4 threads
• 8 GB RAM
THE DATA
Sales: 1,000,004 records
Product: 4,434 records
Customer: 10,000 records
MySQL Query:
SELECT SUM(qty), customer.gender FROM sales JOIN customer ON sales.cid = customer.id GROUP BY customer.gender;
Equivalent BigObject Query:
FIND Customer.gender FROM sales;
Query Time:
588.86 seconds
Query Time:
0.1 seconds
Performance Turbo Boost: 5,888.6x!
Click Here to watch
the Video…
THE ANALYSIS We did not only achieve 1,000x we did it 5 times over! I would never have anticipated we could have achieved a 5,000x performance boost.
I would like to note that we worked with the MySQL tables not being optimized and when they were optimized there was still a 150x improvement on performance. That being said, a lot of databases out there are not optimized so you can expect a turbo-boost of up to 5,000x!
The application initiates a request to the data store to search, insert, update or
delete data.
Application (Web, Mobile, IoT…)
Data Store (MYSQL, POSTGRESQL, NOSQL…)
Application (Web, Mobile, IoT…)
Data Store (MYSQL, POSTGRESQL, NOSQL…)
The datastore replies with a response to the request. Depending on the complexity of the
request, this can take minutes, hours or even days.
Application (Web, Mobile, IoT…)
Data Analysis (R, Python…)
Analysis is performed on data in the data store using data analysis tools.
Data Store (MYSQL, POSTGRESQL, NOSQL…)
Once data analysis has been completed the result is stored in the data store for future
use.
Application (Web, Mobile, IoT…)
Data Analysis (R, Python…)
Data Store (MYSQL, POSTGRESQL, NOSQL…)
The application initiates a request to the data store to search, insert, update or
delete data that seamlessly passes through BigObject.
Application (Web, Mobile, IoT…)
Data Store (MYSQL, POSTGRESQL, NOSQL…)
(invisible pre-cache layer)
BigObject instantly sends back a response and simultaneously sends the request to the data store with an increase of up to 1000x in performance.
If BigObject does not have a response to the request it simply passes the data stores response to the application layer.
Application (Web, Mobile, IoT…)
Data Store (MYSQL, POSTGRESQL, NOSQL…)
(invisible pre-cache layer)
Application (Web, Mobile, IoT…)
Data Store (MYSQL, POSTGRESQL, NOSQL…)
Data Analysis (R, Python…)
BigObject features powerful, built-in real-time data analysis tools. Search, Probability, Statistics & Association.
You may also integrate other data analysis tools seamlessly into BigObject.
(invisible pre-cache layer)
When you store and compute data in the same address space you can achieve massive data analysis performance without the need for massive servers!
“From 2013 to 2020, the digital universe will grow by a factor of 10 – from 4.4 trillion gigabytes to 44 trillion. It more than doubles every two years."
– E M C D I G I TA L U N I V E R S E S T U D Y , 2 0 1 4
– G E O R G E L E E , C I O , I N V E S T M E N T B A N K I N G D I V I S I O N , G O L D M A N S A C H S
“Ninety percent of the world’s data has been created in the last two years… The ultimate question is really what insight and value can we draw from that data.”
With BigObject + Pre-Cache you can now turbo boost your data store with only a few lines of code.
Start performing real-time data analysis and feedback TODAY!
The sky is the limit!
@BIGOBJECT
VISIT WWW.BIGOBJECT.IO FOR VIP ACCESS TODAY!
GET VIP ACCESS!
Credits
Slide 28 Chart http://www.theneweconomy.com/strategy/big-data-is-not-without-its-problems
Slide 26,27,28,29,30,31 Background Photo Joshua Earle
Slides 3,4,5,6,7,9 Background Photo Matthew Wiebe
Slide 2 Photos http://www.twitter.com/titusblair
All other content in this presentation is the copyright of BigObject. All Rights Reserved.
For any assistance with copyright please contact [email protected]