Performance and resource modeling in highly-concurrent OLTP workloads

Database administrators of Online Transaction Processing (OLTP) systems constantly face difcult questions. For example, "What is the maximum throughput I can sustain with my current hardware?", "How much disk I/O will my system perform if the requests per second double?", or &quo...

Full description

Bibliographic Details
Main Authors: Mozafari, Barzan (Author), Curino, Carlo (Author), Jindal, Alekh (Author), Madden, Samuel (Author)
Format: Article
Language:English
Published: ACM Press, 2021-11-09T13:21:13Z.
Subjects:
Online Access:Get fulltext
LEADER 02128 am a22001813u 4500
001 137854
042 |a dc 
100 1 0 |a Mozafari, Barzan  |e author 
700 1 0 |a Curino, Carlo  |e author 
700 1 0 |a Jindal, Alekh  |e author 
700 1 0 |a Madden, Samuel  |e author 
245 0 0 |a Performance and resource modeling in highly-concurrent OLTP workloads 
260 |b ACM Press,   |c 2021-11-09T13:21:13Z. 
856 |z Get fulltext  |u https://hdl.handle.net/1721.1/137854 
520 |a Database administrators of Online Transaction Processing (OLTP) systems constantly face difcult questions. For example, "What is the maximum throughput I can sustain with my current hardware?", "How much disk I/O will my system perform if the requests per second double?", or "What will happen if the ratio of transactions in my system changes?". Resource prediction and performance analysis are both vital and difcult in this setting. Here the challenge is due to high degrees of concurrency, competition for resources, and complex interactions between transactions, all of which non-linearly impact performance. Although difcult, such analysis is a key component in enabling database administrators to understand which queries are eating up the resources, and how their system would scale under load. In this paper, we introduce our framework, called DBSeer, that addresses this problem by employing statistical models that provide resource and performance analysis and prediction for highly concurrent OLTP workloads. Our models are built on a small amount of training data from standard log information collected during normal system operation. Tese models are capable of accurately measuring several performance metrics, including resource consumption on a per-transaction-type basis, resource bottlenecks, and throughput at diferent load levels. We have validated these models on MySQL/Linux with numerous experiments on standard benchmarks (TPC-C) and real workloads (Wikipedia), observing high accuracy (within a few percent error) when predicting all of the above metrics. Copyright © 2013 ACM. 
546 |a en 
655 7 |a Article 
773 |t 10.1145/2463676.2467800