Skip to content

PowerRam ¤

PowerRam()

PowerRam¤

Python classes monitoring RAM's power usage. We used an analytical approach to calibrate RAM power by combining information of DIMM nomminal Power for each DDR memory module version, number of DIMM slots in use and the memory footprint during application execution

PowerRam¤

Python classes monitoring RAM's power usage.

Source code in ea2p/src/ram.py 
20
21
22
23
24
25
26
27
28
29
def __init__(self):
	"""

	PowerRam
	---------------

	Python classes monitoring RAM's power usage.
	"""
	self.ram_power = self.get_memory_power()
	self.number_slots = self.get_number_slots()

append_energy_usage ¤

append_energy_usage()

We used an analytical approach to calibrate RAM power by combining information of DIMM nomminal Power for each DDR memory module version, number of DIMM slots in use and the memory footprint during application execution

Source code in ea2p/src/ram.py 
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
def append_energy_usage(self):
	"""
	We used an analytical approach to calibrate RAM power by combining information of DIMM nomminal Power for each DDR memory module version, number of DIMM slots in use and the memory footprint during application execution
	"""
	ram_usage = psutil.virtual_memory()
	ram_percent = ram_usage[2]
	THRESHOLD0 = 5
	THRESHOLD1 = 10
	THRESHOLD2 = 25
	THRESHOLD3 = 50
	THRESHOLD4 = 70
	if ram_usage[2] <= THRESHOLD0 :
		ram_percent = 35
	elif ram_usage[2] <= THRESHOLD1 :
		ram_percent = 65
	elif ram_usage[2] <= THRESHOLD2 :
		ram_percent = 70
	elif ram_usage[2] <= THRESHOLD3 :
		ram_percent = 75
	elif ram_usage[2] <= THRESHOLD4 :
		ram_percent = 80
	else :
		ram_percent = 85

	energy_usage = {"dram":(self.ram_power * self.number_slots * ram_percent / 100)}
	#print(energy_usage)
	return energy_usage