Download Tobias Mühlbauer, Wolf Rödiger, Angelika Reiser, Alfons Kemper

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts
no text concepts found
Transcript 
ScyPer: A Hybrid OLTP&OLAP Distributed Main Memory
Database System for Scalable Real-Time Analytics
Tobias Mühlbauer, Wolf Rödiger, Angelika Reiser, Alfons Kemper, Thomas Neumann
{muehlbau, roediger, reiser, kemper, neumann}@in.tum.de
Motivation
· Modern hybrid main memory DBMS process > 100,000 TPC-C TX/s in a single HW thread
· Evaluation of OLAP queries on arbitrarily recent snapshots of the transactional database
· OLAP query throughput limited by the number of available hardware threads
⇨ OLTP throughput sufficient, scale-up of OLAP throughput needed
ScyPer Architecture
Row-Store
A B C D E
fork()
Backup
Process
tim
e
OLTP
Shared
Ptn 1
Ptn 2
Ptn 3
Ptn 4
Storage
OLAP
OLTP
Process
Coordinator
Primary HyPer node
Redo Log Multicasting
OLAP Load Balancing
Redo Log
Column-Store
AB
CD
E
Non-Transactional
Data on Disk
OLTP
Process
Ptn 1
Ptn 2
Ptn 3
Ptn 4
Shared
fork()
Scalability of ScyPer
Scalability on a 14 low-end node cluster
e
it m
OLAP
Snapshot
OLAP
Snapshot
Secondary
HyPer node
...
Secondary
HyPer node
Redo Log Multicasting
· PGM-based multicasting
· ~ 700 Mbit/s throughput in 1GbE network
· 100B/log entry ⇨ ~ 900,000 entries per second
Get in touch:
http://www-db.in.tum.de/HyPer
Related documents