Changes between Version 1 and Version 2 of PerformanceData


Ignore:
Timestamp:
Sep 4, 2012 2:48:59 PM (12 years ago)
Author:
Mei
Comment:

--

Legend:

Unmodified
Added
Removed
Modified

  • PerformanceData

    v1 v2  
    1 Performance data
     1[[TOC]]
     2
     3= Performance data =
    24 
    3   We have gathered performance data of two types of ABAC scenario.
     5We have gathered performance data of two types of ABAC scenario.
    46The first is the daisychaining of a large answer proof from a
    57similarly sized ruleset.  The second is the searching of a proof
    68from a noisy ruleset.
    79
    8 ==> daisychain <==
     10== daisychain ==
    911
    10   Many cascading Likes-rules and their matching principal credentials are
     12Many cascading Likes-rules and their matching principal credentials are
    1113generated:
    1214
    13   John0.likes <-John0
    14   John1.likes <-John0
    15   John2.likes <-John1
    16   John3.likes <-John2
    17   ...
    18   JohnMax.likes <-JohnMax-1 
     15{{{
     16John0.likes <- John0
     17John1.likes <- John0
     18John2.likes <- John1
     19John3.likes <- John2
     20..
     21JohnMAX.likes <- JohnMAX-1
     22}}}
    1923
    20   Following graph shows the the time needed to load both attribute
    21 credentials and principal certificates ((2*Max)+2) into the YAP prolog
     24
     25Following graph shows the the time needed to load both attribute
     26credentials and principal certificates ((2*MAX)+2) into the YAP prolog
    2227db. The Y axis is the total time in milliseconds. The X axis is the
    2328number of credentials that are in the YAP backend db.
    … …  
    2631
    2732
    28   A valid query in the form of JohnMax.likes <-?-John0 is issued multiple
     33A valid query,
     34'''JohnMAX.likes<-?-John0'''
     35, is issued multiple
    2936times. The first query's time is collected and then subsequent 10 queries
    3037are collected and an average is taken.  Next graph shows these lines. The
    3138X axis is the query time in microseconds and the Y axis is number of
    3239credentials in the YAP backend db.
     40
    3341 
    3442XXX
    3543
    36   An invalid query in the form of John0.likes <-?-JohnMax is issued
     44An invalid query in the form of '''John0.likes<-?-JohnMAX''' is issued
    3745multiple times. The first query's time is collected and then subsequent
    384610 queries are collected and an average is taken.  Next graph shows these
    … …  
    4351
    4452
    45 ==> haystack <==
     53== haystack ==
    4654
    47   The core credential setup is borrowed from the
     55The core credential setup is borrowed from the
    4856Ralphs' fruitprice and shopper's eating preference RT2 example,
    4957 
    50   Mary.what2eat <- Ralphs.fruitprice(P:[.. 2.00])
    51   Bob.what2eat <- Ralphs.fruitprice(P:[1.00 .. 5.00])
    52   Ralphs.fruitprice(1.50) <- 'apple'
    53   Ralphs.fruitprice(1.50) <- 'kiwi'
    54   Ralphs.fruitprice(2.50) <- 'black rasberry'
    55   Ralphs.fruitprice(0.50) <- 'navel orange'
     58{{{
     59Mary.what2eat <- Ralphs.fruitprice(P:[..2.00])
     60Bob.what2eat <- Ralphs.fruitprice(P:[1.00 .. 5.00])
     61Ralphs.fruitprice(1.50) <- 'apple'
     62Ralphs.fruitprice(1.50) <- 'kiwi'
     63Ralphs.fruitprice(2.50) <- 'black rasberry'
     64Ralphs.fruitprice(0.50) <- 'navel orange'
     65}}}
    5666
    57   Noises are introduced by adding the rules about bananas
     67Noises are introduced by adding the rules about bananas
    5868and Johns.
    5969
    60   Ralphs.fruitprice(1.00) <- 'banana1'
    61   Ralphs.fruitprice(2.00) <- 'banana2'
    62   ..
    63   Ralphs.fruitprice(Max.00) <- bananaMax
     70{{{
     71Ralphs.fruitprice(1.00) <- 'banana1'
     72Ralphs.fruitprice(2.00) <- 'banana2'
     73..
     74Ralphs.fruitprice(MAX.00) <- 'bananaMAX'
     75}}}
    6476
    65   John1.what2eat <- Ralphs.fruitprice(P:[1.00 .. 5.00])
    66   John2.what2eat <- Ralphs.fruitprice(P:[1.00 .. 5.00])
    67   ..
    68   JohnMAX.what2eat <- Ralphs.fruitprice(P:[1.00 .. 5.00])
     77and,
    6978
    70   Following graph shows the the time needed to load both attribute
    71 credentials and principal certificates ((3*Max)+9) into the YAP prolog
     79{{{
     80John1.what2eat <- Ralphs.fruitprice(P:[1.00 .. 5.00])
     81John2.what2eat <- Ralphs.fruitprice(P:[1.00 .. 5.00])
     82..
     83JohnMAX.what2eat <- Ralphs.fruitprice(P:[1.00 .. 5.00])
     84}}}
     85
     86Following graph shows the the time needed to load both attribute
     87credentials and principal certificates ((3*MAX)+9) into the YAP prolog
    7288db. The Y axis is the total time in milliseconds. The X axis is the
    7389number of credentials that are in the YAP backend db.
    … …  
    7591XXX
    7692
    77   A valid query, Mary.what2eat<-?-'navel orange', is issued multiple
     93A valid query, '''Mary.what2eat<-?-'navel orange'''', is issued multiple
    7894times. The returning answer proof is,
    79      Mary.what2eat <- pRalphs.fruitprice(P:[..2.000000])
    80      Ralphs.fruitprice(0.50) <- 'navel orange'. 
     95{{{
     96Mary.what2eat <- pRalphs.fruitprice(P:[..2.00])
     97Ralphs.fruitprice(0.50) <- 'navel orange'
     98}}}
     99 
    81100The first query's time is collected and the subsequent 20 queries
    82101are collected and averaged. Next graph shows these lines. The X
    … …  
    84103microseconds and the Y axis is number of credentials in the YAP
    85104backend db.
    86  
     105
    87106XXX
    88107
    89   An invalid query in the form of Bob.what2eat<-?-'navel orange'
     108An invalid query in the form of '''Bob.what2eat<-?-'navel orange''''
    90109is issued multiple times. The first query's time is collected and
    91110then subsequent 20 queries are collected and averaged.  Next graph
    92111shows these lines. The X axis is the query time in microseconds
    93 and the Y axis is number of credentialsi in the YAP backend db.
     112and the Y axis is number of credentials in the YAP backend db.
    94113
    95114XXX