Mean Value Analysis of a Database Grid Application - PowerPoint PPT Presentation

Loading...

PPT – Mean Value Analysis of a Database Grid Application PowerPoint presentation | free to download - id: 6b0dd5-ZWE1Z



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Mean Value Analysis of a Database Grid Application

Description:

Mean Value Analysis of a Database Grid Application Dale R. Thompson Computer Science and Computer Engineering University of Arkansas – PowerPoint PPT presentation

Number of Views:21
Avg rating:3.0/5.0
Date added: 4 June 2020
Slides: 20
Provided by: DaleRTh
Learn more at: http://www.csce.uark.edu
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Mean Value Analysis of a Database Grid Application


1
Mean Value Analysis of a Database Grid Application
  • Dale R. Thompson
  • Computer Science and Computer Engineering
  • University of Arkansas

2
Introduction
  • The analysis of a queueing network is important
    for predicting the performance of a system.
  • A database grid application was modeled using a
    queueing network.
  • The queueing network was analyzed by using an
    approximate mean value analysis algorithm called
    the Bard-Schweitzer algorithm or the proportional
    estimate (PE) algorithm.
  • Several different types of record flows were
    modeled. For example, uniform, non-uniform, etc.
  • A system in which the batch and interactive
    requests are segregated was modeled.

3
Queueing Network System
4
MVA algorithms
when class c is a batch processing, Zc0
  • Mean Value Analysis calculates throughput (Xc),
    response time (Rc), and queue length (Qd,c)of
    each class.
  • It can be classified with the number of client
    open or closed.
  • It also can be classified with how to get the
    values Exactly or Approximately.
  • Single class or multiple classes?


5
Classification of MVA algorithms
6
Comparison of MVA algorithms
Algorithms Rank of Accuracy Space Complexity Time Complexity
Exact MVA 1 O(KC?Cc1(Nc1) O(KC?Cc1(Nc1)
Linearizer 2 O(KC2)/iteration O(KC3)/iteration
PE 3 O(KC)/iteration O(KC)/iteration
LCP 4 O(KC)/iteration O(KC)/iteration
Algorithms Rank of Accuracy Rank of Accuracy Space Complexity Time Complexity
Algorithms No con. Congestion Space Complexity Time Complexity
Exact MVA 1 1 O(KC?Cc1(Nc1) O(KC?Cc1(Nc1)
FL 2 4 O(KC)/iteration O(KC)/iteration
QL 3 2 O(KC)/iteration O(KC)/iteration
PE 4 3 O(KC)/iteration O(KC)/iteration
7
Database Grid Application
High-level Overview of System
Database Link Application Example
8
Database Grid Application Cont.
A high-level view of the grid
The Flow of Records
9
Current Queueing Model
Queueing Model
10
CPU and Network Demand
  • Record size - 500bytes, Ethernet - 26bytes, IP -
    20bytes, TCP - 20bytes. Total actual record size
    - 566bytes
  • Service demand1 computers in the clients, the
    director grid, and database grid.
  • Service demand2 network cards in the clients,
    the director grid, and database grid.

11
Maximum Throughput and Block size
  • Maximum attainable throughput 79.5Mega
    record/hr
  • The block size
  • Batch class 1150 records
  • Interactive class 1 record.

12
Uniforms Distributions
  • Each record was equally likely to go to any of
    the computers in the database grid,
  • Block size varying

13
Non-uniform Distribution
  • Non-uniform distribution of demands was created
    by assuming that
  • 20 clients 10,15,20 director computers 70
    database computers
  • 80 of the requests from clients (16 clients) gt
    20 of the database grid (14 Computers).
  • The remaining 20 of the requests (4 clients)gt
    the remaining 80 of the database grid (56
    Computers)

Throughput (Mrecords/hr) 10 directors 15 directors 20 directors
Uniform 40 59 79
Non-uniform 40 59 71
Delay time (s/records) 10 directors 15 directors 20 directors
Uniform 0.1043 0.0783 0.0523
Non-uniform 0.1043 0.0783 0.0581
14
Uniform Number of Clients
  • Uniform 1150 records
  • Varying number of Clients 20, 40, 60

Throughput (Mrecords/hr) 20 clients 40 clients 60 clients
10 directors 40 20 13
15 directors 59 30 20
20 directors 79 40 26
Delay time (s/record) 20 clients 40 clients 60 clients
10 directors 0.1043 0.2084 0.3126
15 directors 0.0783 0.1433 0.2084
20 directors 0.0523 0.1043 0.1564
15
Proposed Change to application
  • It was assumed that there were two updates per
    request
  • This proposed change was modeled by having 5 of
    the clients (1 client out of 20) require demand
    from two different database grid computers.
  • Block size 1150 records

Throughput (Mrecords/hr) 10 directors 15 directors 20 directors
Original Application 40 59 79
Proposed Application 39 57 77
decrease 2.50 3.32 2.49
Delay time (s/record) 10 directors 15 directors 20 directors
Original Application 0.1043 0.0783 0.0523
Proposed Application 0.1095 0.0809 0.0549
increase 4.98 3.32 4.97
16
Segregation of batch and interactive classes
  • This model is for the actual system.
  • 20 clients 16 director computers 70 database
    computers.
  • There are 12 clients batch and 8 clients
    interactive record.
  • Batch 12 clients gt 12 directors
  • Interactive 8 clients gt 4 directors.

Throughput (Mrecords/hr)
Interactive 3.4
Batch 47.5
Total 50.9
Avg. Delay (s/record)
Interactive 0.0002
Batch 0.0314
Total 0.0316
  • This reduces the mean delay per record to better
    serve the interactive clients
  • The database link application could use the
    0.0002 s/record parameter as a design parameter

17
Conclusions
  • The number of directors should be approximately
    equal to the number of clients to obtain the
    maximum throughput of the system.
  • The bottleneck device in this system is the
    network.
  • The proposed application change that caused 5 of
    the records to require service from two database
    grid computers did not significantly decrease the
    performance of the system.
  • Segregating the batch and interactive classes at
    the director level causes the response time of
    the interactive classes to decrease. The
    decreased response time comes at the price of
    lowering the overall throughput of the system. As
    discussed, the model can be used to determine the
    trade offs of decreased response time versus
    increased throughput.

18
Future Work
  • Traffic analysis of submitted records
  • Simulation of alternate configurations
  • Scheduling of grid computers
  • Modeling/Simulation of different applications
  • Grid-enable applications that run in different
    locations and organizations
  • Others?

19
Contact Information and Copy of this Presentation
  • Dale R. Thompson
  • 311 Engineering Hall
  • Fayetteville, Arkansas, USA
  • 72701
  • Phone 1 (479) 575-5090
  • E-mail drt_at_uark.edu
  • WWW http//csce.uark.edu/drt
About PowerShow.com