cellinfo
/
cocluto
2
1
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

fixed pep8 styling problems

This commit is contained in:
Guillaume Raffy 2021-02-23 13:01:52 +00:00
parent 78d7e02285
commit 14e3e12571
1 changed files with 71 additions and 61 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

132
cluster_stats.py
View File

@ -1,5 +1,5 @@
# encoding: utf-8 # encoding: utf-8
import sys # import sys
import os import os
import re import re
import datetime import datetime
@ -16,47 +16,51 @@ import matplotlib
# https://matplotlib.org/tutorials/introductory/usage.html#what-is-a-backend # https://matplotlib.org/tutorials/introductory/usage.html#what-is-a-backend
r = os.system('python -c "import matplotlib.pyplot as plt;plt.figure()"') r = os.system('python -c "import matplotlib.pyplot as plt;plt.figure()"')
if r != 0: if r != 0:
matplotlib.use('Agg') # use Anti Grain Geometry backend for non-interactive rendering into pngs, svg, etc... matplotlib.use('Agg') # use Anti Grain Geometry backend for non-interactive rendering into pngs, svg, etc...
import matplotlib.pyplot as plt import matplotlib.pyplot as plt
#import matplotlib.pyplot as plt # import matplotlib.pyplot as plt
import matplotlib.dates import matplotlib.dates
import abc import abc
from SimpaDbUtil import SqlDatabaseReader, SqlFile # from SimpaDbUtil import SqlDatabaseReader, SqlFile
from inventory import Inventory # from inventory import Inventory
def is_cluster_node_name(name): def is_cluster_node_name(name):
return re.match('^simpatix[0-9]+$', name) is not None or re.match('^physix[0-9]+$', name) is not None return re.match('^simpatix[0-9]+$', name) is not None or re.match('^physix[0-9]+$', name) is not None
def get_investment_over_time(time_value, price, purchase_time): def get_investment_over_time(time_value, price, purchase_time):
percent_decay_per_day = 0.0 # 1.0/(7.0*365.0) percent_decay_per_day = 0.0 # 1.0/(7.0*365.0)
f1 = (purchase_time-time_value)*percent_decay_per_day+1.0 f1 = (purchase_time - time_value) * percent_decay_per_day + 1.0
f2 = np.where( f1 < 0.0, 0.0, f1 ) f2 = np.where(f1 < 0.0, 0.0, f1)
f3 = np.where( time_value < purchase_time, 0.0, f2 ) f3 = np.where(time_value < purchase_time, 0.0, f2)
return f3 * price return f3 * price
def get_flops_over_time(inventory, time_value, computer_id, purchase_time): def get_flops_over_time(inventory, time_value, computer_id, purchase_time):
""" """
:param Inventory inventory: :param Inventory inventory:
""" """
return np.where( time_value < purchase_time, 0.0, inventory.get_computer_dflops(computer_id) ) return np.where(time_value < purchase_time, 0.0, inventory.get_computer_dflops(computer_id))
def get_flops_price_over_time(inventory, time_value): def get_flops_price_over_time(inventory, time_value):
""" """
:param Inventory inventory: the inventory database :param Inventory inventory: the inventory database
""" """
rows = inventory.query("SELECT * FROM machines") rows = inventory.query("SELECT * FROM machines")
def get_key(item): def get_key(item):
return item['time'] return item['time']
flops_prices = [] flops_prices = []
for row in rows: for row in rows:
(name, serial_number, affectation, machine_spec_id, command_id, price_ex_vat, pos_x, pos_y, pos_z, inv_number)=row (name, serial_number, affectation, machine_spec_id, command_id, price_ex_vat, pos_x, pos_y, pos_z, inv_number) = row
is_cluster_node = is_cluster_node_name(name) is_cluster_node = is_cluster_node_name(name)
if is_cluster_node: if is_cluster_node:
purchase_date = inventory.get_machine_purchase_date(name) purchase_date = inventory.get_machine_purchase_date(name)
@ -64,20 +68,20 @@ def get_flops_price_over_time(inventory, time_value):
# print(name, price_ex_vat) # print(name, price_ex_vat)
purchase_time = matplotlib.dates.date2num(purchase_date.date()) purchase_time = matplotlib.dates.date2num(purchase_date.date())
computer_flops = inventory.get_computer_dflops(name) computer_flops = inventory.get_computer_dflops(name)
flops_price = ( price_ex_vat-inventory.get_computer_options_price(name) ) / computer_flops flops_price = (price_ex_vat - inventory.get_computer_options_price(name)) / computer_flops
# print ( purchase_date, name, price_ex_vat, computer_flops, flops_price ) # print ( purchase_date, name, price_ex_vat, computer_flops, flops_price )
flops_prices.append({'time':purchase_time, 'flops_price':flops_price, 'purchase_date':purchase_date}) flops_prices.append({'time': purchase_time, 'flops_price': flops_price, 'purchase_date': purchase_date})
flops_prices = sorted(flops_prices, key=get_key) flops_prices = sorted(flops_prices, key=get_key)
flops_price_over_time = np.where( True, 0.0, 0.0 ) flops_price_over_time = np.where(True, 0.0, 0.0)
for item in flops_prices: for item in flops_prices:
# print(item) # print(item)
flops_price_over_time = np.where( time_value < item['time'], flops_price_over_time, item['flops_price']) flops_price_over_time = np.where(time_value < item['time'], flops_price_over_time, item['flops_price'])
return flops_price_over_time return flops_price_over_time
def get_computer_value_over_time(inventory, computer_id, time_value, flops_price_over_time, purchase_time): def get_computer_value_over_time(inventory, computer_id, time_value, flops_price_over_time, purchase_time):
# print('flops_price_over_time = ', flops_price_over_time) # print('flops_price_over_time = ', flops_price_over_time)
computer_flops = inventory.get_computer_dflops(computer_id) computer_flops = inventory.get_computer_dflops(computer_id)
@ -98,9 +102,9 @@ def get_computer_value_over_time(inventory, computer_id, time_value, flops_price
# plt.legend(list(y_signals.keys())) # plt.legend(list(y_signals.keys()))
# else: # else:
# colors = ['blue', 'orange', 'green', 'purple', 'yellow'] # colors = ['blue', 'orange', 'green', 'purple', 'yellow']
# # emulating missing Axes.stackplot method # # emulating missing Axes.stackplot method
# y = np.row_stack(list(y_signals.itervalues())) # y = np.row_stack(list(y_signals.itervalues()))
# # this call to 'cumsum' (cumulative sum), passing in your y data, # # this call to 'cumsum' (cumulative sum), passing in your y data,
# # is necessary to avoid having to manually order the datasets # # is necessary to avoid having to manually order the datasets
# y_stack = np.cumsum(y, axis=0) # a 3x10 array # y_stack = np.cumsum(y, axis=0) # a 3x10 array
# for series_index in range(len(y_signals)): # for series_index in range(len(y_signals)):
@ -110,6 +114,8 @@ def get_computer_value_over_time(inventory, computer_id, time_value, flops_price
# from_signal = y_stack[series_index-1,:] # from_signal = y_stack[series_index-1,:]
# ax.fill_between(x_signal, from_signal, y_stack[series_index,:], color=colors[series_index], lw=0.0, label=y_signals.keys()[series_index]) # ax.fill_between(x_signal, from_signal, y_stack[series_index,:], color=colors[series_index], lw=0.0, label=y_signals.keys()[series_index])
# plt.legend() # plt.legend()
def stackplot(ax, x_signal, y_signals): def stackplot(ax, x_signal, y_signals):
""" """
:param matplotlib.Axes ax: :param matplotlib.Axes ax:
@ -117,7 +123,7 @@ def stackplot(ax, x_signal, y_signals):
:param dict(str,numpy.array) y_signals: :param dict(str,numpy.array) y_signals:
""" """
if 'stackplot' in dir(ax): if 'stackplot' in dir(ax):
ax.stackplot(x_signal, list(y_signals.values()) ) ax.stackplot(x_signal, list(y_signals.values()))
plt.legend(list(y_signals.keys())) plt.legend(list(y_signals.keys()))
else: else:
# emulating missing Axes.stackplot method # emulating missing Axes.stackplot method
@ -130,21 +136,22 @@ def stackplot(ax, x_signal, y_signals):
if series_index == 0: if series_index == 0:
from_signal = 0 from_signal = 0
else: else:
from_signal = y_stack[series_index-1,:] from_signal = y_stack[series_index - 1, :]
ax.fill_between(x_signal, from_signal, y_stack[series_index,:], color=colors[series_index], lw=0.0) ax.fill_between(x_signal, from_signal, y_stack[series_index, :], color=colors[series_index], lw=0.0)
p = plt.Rectangle((0, 0), 0, 0, color=colors[series_index]) p = plt.Rectangle((0, 0), 0, 0, color=colors[series_index])
ax.add_patch(p) ax.add_patch(p)
plt.legend(list(y_signals.keys())) plt.legend(list(y_signals.keys()))
def draw_cluster_value_over_time_graph(inventory, from_date, to_date, graph_type): def draw_cluster_value_over_time_graph(inventory, from_date, to_date, graph_type):
time_value = matplotlib.dates.drange(dstart=from_date, dend=to_date, delta=datetime.timedelta(days=1)) time_value = matplotlib.dates.drange(dstart=from_date, dend=to_date, delta=datetime.timedelta(days=1))
flops_price_over_time = get_flops_price_over_time(inventory, time_value) flops_price_over_time = get_flops_price_over_time(inventory, time_value)
cluster_value = {} cluster_value = {}
rows = inventory.query("SELECT * FROM machines") rows = inventory.query("SELECT * FROM machines")
for row in rows: for row in rows:
(name, serial_number, affectation, machine_spec_id, command_id, price_ex_vat, pos_x, pos_y, pos_z, inv_number)=row (name, serial_number, affectation, machine_spec_id, command_id, price_ex_vat, pos_x, pos_y, pos_z, inv_number) = row
is_cluster_node = is_cluster_node_name(name) is_cluster_node = is_cluster_node_name(name)
if is_cluster_node: if is_cluster_node:
purchase_date = inventory.get_machine_purchase_date(name) purchase_date = inventory.get_machine_purchase_date(name)
@ -152,9 +159,9 @@ def draw_cluster_value_over_time_graph(inventory, from_date, to_date, graph_type
# print(name, price_ex_vat) # print(name, price_ex_vat)
purchase_time = matplotlib.dates.date2num(purchase_date.date()) purchase_time = matplotlib.dates.date2num(purchase_date.date())
item_value_over_time = { item_value_over_time = {
'cluster_cost_over_time':get_investment_over_time(time_value, price_ex_vat, purchase_time), 'cluster_cost_over_time': get_investment_over_time(time_value, price_ex_vat, purchase_time),
'cluster_value_over_time':get_computer_value_over_time(inventory, name, time_value, flops_price_over_time, purchase_time), 'cluster_value_over_time': get_computer_value_over_time(inventory, name, time_value, flops_price_over_time, purchase_time),
'cluster_dp_gflops_over_time':get_flops_over_time(inventory, time_value, name, purchase_time)}[graph_type] 'cluster_dp_gflops_over_time': get_flops_over_time(inventory, time_value, name, purchase_time)}[graph_type]
for ownership in inventory.get_item_ownership(name): for ownership in inventory.get_item_ownership(name):
# print(ownership) # print(ownership)
# print(ownership['owner'], ownership['owner_ratio']) # print(ownership['owner'], ownership['owner_ratio'])
@ -165,7 +172,7 @@ def draw_cluster_value_over_time_graph(inventory, from_date, to_date, graph_type
if owner_dept in cluster_value.keys(): if owner_dept in cluster_value.keys():
cluster_value[owner_dept] += item_value_over_time cluster_value[owner_dept] += item_value_over_time
else: else:
cluster_value[owner_dept] = item_value_over_time # np.zeros_like(time_value) cluster_value[owner_dept] = item_value_over_time # np.zeros_like(time_value)
# print(purchase_date) # print(purchase_date)
# print(type(from_date)) # print(type(from_date))
@ -175,17 +182,17 @@ def draw_cluster_value_over_time_graph(inventory, from_date, to_date, graph_type
fig, ax = plt.subplots() fig, ax = plt.subplots()
ax.set_title(graph_type) ax.set_title(graph_type)
#for dept, cluster_value_for_dept in cluster_value.iteritems(): # for dept, cluster_value_for_dept in cluster_value.iteritems():
# ax.plot(time_value, cluster_value_for_dept) # ax.plot(time_value, cluster_value_for_dept)
stackplot( ax, time_value, cluster_value) stackplot(ax, time_value, cluster_value)
plt.xlabel('time') plt.xlabel('time')
plt.ylabel( plt.ylabel({
{'cluster_cost_over_time':u'cluster investment (€)', 'cluster_cost_over_time': u'cluster investment (€)',
'cluster_value_over_time':u'cluster value (€)', 'cluster_value_over_time': u'cluster value (€)',
'cluster_dp_gflops_over_time':u'double prec gflops'}[graph_type]) 'cluster_dp_gflops_over_time': u'double prec gflops'}[graph_type])
years = matplotlib.dates.YearLocator() # every year years = matplotlib.dates.YearLocator() # every year
months = matplotlib.dates.MonthLocator() # every month months = matplotlib.dates.MonthLocator() # every month
yearsFmt = matplotlib.dates.DateFormatter('%Y') yearsFmt = matplotlib.dates.DateFormatter('%Y')
@ -194,7 +201,7 @@ def draw_cluster_value_over_time_graph(inventory, from_date, to_date, graph_type
ax.xaxis.set_major_locator(years) ax.xaxis.set_major_locator(years)
ax.xaxis.set_major_formatter(yearsFmt) ax.xaxis.set_major_formatter(yearsFmt)
ax.xaxis.set_minor_locator(months) ax.xaxis.set_minor_locator(months)
datemin = datetime.date(from_date.year, 1, 1) datemin = datetime.date(from_date.year, 1, 1)
datemax = datetime.date(to_date.year + 1, 1, 1) datemax = datetime.date(to_date.year + 1, 1, 1)
ax.set_xlim(datemin, datemax) ax.set_xlim(datemin, datemax)
@ -203,18 +210,19 @@ def draw_cluster_value_over_time_graph(inventory, from_date, to_date, graph_type
# axes up to make room for them # axes up to make room for them
# fig.autofmt_xdate() # fig.autofmt_xdate()
ax.grid(True) ax.grid(True)
#plt.plot() # plt.plot()
# plt.plot(X,S) # plt.plot(X,S)
return fig return fig
def draw_dp_gflops_price_over_time_over_time_graph(inventory, from_date, to_date): def draw_dp_gflops_price_over_time_over_time_graph(inventory, from_date, to_date):
""" """
:param Inventory inventory: the inventory database :param Inventory inventory: the inventory database
:param datetime from_time: :param datetime from_time:
:param datetime to_time: :param datetime to_time:
""" """
time_value = matplotlib.dates.drange(dstart=from_date, dend=to_date, delta=datetime.timedelta(days=1)) time_value = matplotlib.dates.drange(dstart=from_date, dend=to_date, delta=datetime.timedelta(days=1))
gflops_price_over_time = get_flops_price_over_time(inventory, time_value) * 1.0e9 gflops_price_over_time = get_flops_price_over_time(inventory, time_value) * 1.0e9
fig, ax = plt.subplots() fig, ax = plt.subplots()
@ -223,7 +231,7 @@ def draw_dp_gflops_price_over_time_over_time_graph(inventory, from_date, to_date
ax.set_xlabel('time') ax.set_xlabel('time')
ax.set_ylabel(u'double precision flops price (€/gflops)') ax.set_ylabel(u'double precision flops price (€/gflops)')
ax.set_title('gflops_price_over_time') ax.set_title('gflops_price_over_time')
years = matplotlib.dates.YearLocator() # every year years = matplotlib.dates.YearLocator() # every year
months = matplotlib.dates.MonthLocator() # every month months = matplotlib.dates.MonthLocator() # every month
yearsFmt = matplotlib.dates.DateFormatter('%Y') yearsFmt = matplotlib.dates.DateFormatter('%Y')
@ -235,6 +243,7 @@ def draw_dp_gflops_price_over_time_over_time_graph(inventory, from_date, to_date
ax.grid(True) ax.grid(True)
return fig return fig
def draw_age_pyramid_graph(inventory): def draw_age_pyramid_graph(inventory):
""" """
:param Inventory inventory: the inventory database :param Inventory inventory: the inventory database
@ -246,12 +255,12 @@ def draw_age_pyramid_graph(inventory):
rows = inventory.query("SELECT * FROM machines") rows = inventory.query("SELECT * FROM machines")
for row in rows: for row in rows:
(name, serial_number, affectation, machine_spec_id, command_id, price_ex_vat, pos_x, pos_y, pos_z, inv_number)=row (name, serial_number, affectation, machine_spec_id, command_id, price_ex_vat, pos_x, pos_y, pos_z, inv_number) = row
is_cluster_node = is_cluster_node_name(name) is_cluster_node = is_cluster_node_name(name)
if is_cluster_node: if is_cluster_node:
purchase_date = inventory.get_machine_purchase_date(name) purchase_date = inventory.get_machine_purchase_date(name)
if purchase_date is not None: if purchase_date is not None:
purchase_time = matplotlib.dates.date2num(purchase_date.date()) purchase_time = matplotlib.dates.date2num(purchase_date.date()) # noqa: F841
age = datetime.datetime.now() - purchase_date age = datetime.datetime.now() - purchase_date
age_histogram[age.days / 365] += 1 age_histogram[age.days / 365] += 1
# print(name, age) # print(name, age)
@ -260,10 +269,10 @@ def draw_age_pyramid_graph(inventory):
ax.bar(range(oldest_age), age_histogram) ax.bar(range(oldest_age), age_histogram)
ax.set_xlabel('age (in years)') ax.set_xlabel('age (in years)')
ax.set_xticks(range(oldest_age)) ax.set_xticks(range(oldest_age))
ax.set_ylabel(u'number of compute nodes') ax.set_ylabel(u'number of compute nodes')
ax.set_title('compute_nodes_age_pyramid') ax.set_title('compute_nodes_age_pyramid')
# format the ticks # format the ticks
ax.grid(True) ax.grid(True)
return fig return fig
@ -273,14 +282,14 @@ class IFigureHandler(object):
""" """
specifies what to do with generated figures specifies what to do with generated figures
""" """
@abc.abstractmethod @abc.abstractmethod
def on_figure_ended(self, fig): def on_figure_ended(self, fig):
""" """
:param matplotlib.Figure fig: :param matplotlib.Figure fig:
""" """
pass pass
@abc.abstractmethod @abc.abstractmethod
def on_finalize(self): def on_finalize(self):
""" """
@ -288,44 +297,45 @@ class IFigureHandler(object):
""" """
pass pass
class ScreenFigureHandler(IFigureHandler): class ScreenFigureHandler(IFigureHandler):
""" """
displays figures on screen displays figures on screen
""" """
def __init__(self): def __init__(self):
pass pass
def on_figure_ended(self, fig): def on_figure_ended(self, fig):
pass pass
def on_finalize(self): def on_finalize(self):
plt.show() plt.show()
class SvgFigureHandler(IFigureHandler): class SvgFigureHandler(IFigureHandler):
""" """
saves figures as svg files saves figures as svg files
""" """
def __init__(self):
pass
def __init__(self, out_svg_dir_path): def __init__(self, out_svg_dir_path):
""" """
:param str out_svg_dir_path: where to save the svg files :param str out_svg_dir_path: where to save the svg files
""" """
self._out_svg_dir_path = out_svg_dir_path self._out_svg_dir_path = out_svg_dir_path
def on_figure_ended(self, fig): def on_figure_ended(self, fig):
fig.savefig(self._out_svg_dir_path + '/' + fig.axes[0].get_title() + '.svg') fig.savefig(self._out_svg_dir_path + '/' + fig.axes[0].get_title() + '.svg')
def on_finalize(self): def on_finalize(self):
pass pass
def draw_graphs(inventory, from_time, to_time, figure_handler): def draw_graphs(inventory, from_time, to_time, figure_handler):
""" """
:param Inventory inventory: the inventory database :param Inventory inventory: the inventory database
:param datetime from_time: :param datetime from_time:
:param datetime to_time: :param datetime to_time:
:param IFigureHandler figure_handler: :param IFigureHandler figure_handler:
""" """
fig = draw_cluster_value_over_time_graph(inventory, from_time.date(), to_time.date(), 'cluster_value_over_time') fig = draw_cluster_value_over_time_graph(inventory, from_time.date(), to_time.date(), 'cluster_value_over_time')
figure_handler.on_figure_ended(fig) figure_handler.on_figure_ended(fig)