Your SlideShare is downloading. ×
Memsql product overview_2013
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×

Introducing the official SlideShare app

Stunning, full-screen experience for iPhone and Android

Text the download link to your phone

Standard text messaging rates apply

Memsql product overview_2013

558
views

Published on


0 Comments
2 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total Views
558
On Slideshare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
0
Comments
0
Likes
2
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide

Transcript

  • 1. PRODUCT OVERVIEW AND ROADMAPJanuary 10, 2013
  • 2. ABOUT US.FOUNDERS COMPANY Founded: 2011 Eric Frenkiel, CEO 15 employees Based in San Francisco, New York INVESTORS Nikita Shamgunov, CTO Paul Buccheit, creator of Gmail Aaron Levie, CEO of Box Max Levchin, cofounder of Paypal
  • 3. OUR VISION: SPEED AT SCALE.Big Data is only useful if it’s accessibleSQL is a must-have for analyticsScale must be achieved on commodity hardware
  • 4. PRODUCT.MemSQL is an ANSI SQL-compliant, distributed, in-memory database
  • 5. REAL-TIME + HISTORICAL DATA.Time value of dataExamples: clicks, tickdata, machine data,impressions T -4 T0 3 2 1 T -1 T0 3 2 T -2 T0 1 T -1 T0 T0Difficult to queryreal-time andhistorical datasimultaneously
  • 6. REAL-TIME ANALYTICS.MemSQL can consume millions of events/secondIssue concurrent OLAP queries via SQL interface againstlock-free data structuresFlexible and interoperable with existing toolsSub-millisecond response times on terabyte-scaleworkloads
  • 7. WHY MEMORY MATTERS.Memory is faster than diskData is an inflationary coefficient in the enterpriseCPU is going massively multi-coreStorage, especially memory, is dropping
  • 8. TECHNOLOGY INNOVATIONS.
  • 9. EXECUTION ENGINE.Execute in parallel across the clusteron every coreSqueeze each CPU cycle using SQL-to-machine code conversion for efficientexecutionAuto-parameterization keepscompilation to a minimum
  • 10. STORAGE ENGINE.Remove locks from the systemNo need for B-trees: use lock-free skip listsand hash tables for massive concurrencyMVCCACID compliant
  • 11. DISTRIBUTED SYSTEM.Shared-nothing architectureDistributed query optimizerHighly available, fault tolerant Client ... Client Aggregator Aggregator Aggregator MemSQL leaf MemSQL leaf ... MemSQL leaf MemSQL leaf
  • 12. PROCESS IN PARALLEL.The goal: leverage hundreds/thousands of coressimultaneously.A distributed system to manage state isfundamental for successful infrastructure.Hadoop works, but it’s Not real-time Hard to use for complex querying
  • 13. MODERN PROBLEMS, MODERN SOLUTIONS.Scale horizontally on commodity hardwareRAM is the new disk; 100+ core clusters are thenormIt’s good to be impatient: real-time analytics
  • 14. THANK YOU.CONTACT WEB 425 2nd STsales@memsql.com www.memsql.com San Francisco, CA 94107 200 Park S Ave New York, NY 10003