https://github.com/jhb/zodbtime
measurements on zodb and persistent objects
https://github.com/jhb/zodbtime
Last synced: 3 months ago
JSON representation
measurements on zodb and persistent objects
- Host: GitHub
- URL: https://github.com/jhb/zodbtime
- Owner: jhb
- License: gpl-2.0
- Created: 2013-08-13T12:02:13.000Z (almost 12 years ago)
- Default Branch: master
- Last Pushed: 2013-08-16T23:42:18.000Z (almost 12 years ago)
- Last Synced: 2025-02-24T07:58:57.847Z (3 months ago)
- Language: Python
- Homepage:
- Size: 156 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
zodbtime
========
Measurements on zodb and persistent objects
All the code creates 100k nodes and 1m edge objects. Whats differing is the format of the edges, or their content.
Some extract from the code is:
g=root['graphdb']
g.nodes=IOBTree()
g.edges=IOBTree()
#...
for j in range(1,1000001):
edge = [[j,random.randint(1,100000),random.randint(1,100000),{}]
g.edges[j]=edge
#...
# loading
x = list(g.edges.values())
# extracting
y = [str(e[3]) for e in x]
The differences between the tests:
Test edge node refs data
==== ==== ========= ====
testid: list int empty dict
testplist: PersistentList int empty dict
testref: list object ref empty dict
testpref: PersistentList object ref empty dict
testdatatuple: list int tuple
testdataplist: list int PersistentList
testdatadict: list int dict
testdatapmap: list int PersistentMapping
testdataclass: list int 'normal' class instance
testdatapclass: list int Persistence inheriting class instance
The results:
Test Data.fs (MB) Init RAM (MB) load time(s) loaded RAM(MB) extract time(s) extracted RAM (MB)
==== ============ ============= ============ ============== =============== ==================
testid 30 181 21 2.28 692 531 0.09 693 533
testplist 133 189 29 3.593 432 271 28.3 1239 1100
testref 38 181 21 3.349 1081 920 0.13 1082 921
testpref 141 189 29 3.583 432 271 30.69 1726 1500
testdatatuple 30 181 21 1.883 422 261 0.04 423 262
testdataplist 148 189 29 4.697 636 476 0.06 652 483
testdatadict 35 181 21 2.388 694 533 0.09 695 534
testpmap 153 189 29 4.617 636 474 0.06 650 483
testdataclass 49 181 21 3.994 756 595 0.08 757 596
testdatapclass 121 189 29 4.409 636 476 0.06 652 483
The order of steps was the following:
- create Data with `python .py`
- ls -alh Data.fs Data.fs (MB)
- open console with `ipython -i console.py`
- look in `top` to see ram values Init RAM (MB)
- execute `timeit()` in console (this loads all the edges) load time(s)
- look in `top` to see ram values loaded RAM(MB)
- execute `extract()` in console (this accesses the last data element in the edges) extract time(s)
- look in `top` to see ram values extracted RAM (MB)