mirror of
https://github.com/logos-storage/das-research.git
synced 2026-01-08 08:03:10 +00:00
Considering Node based model, not validator based
This commit is contained in:
Sudipta Basak
parent
0656789c12
commit
74c003dd72
@ -26,17 +26,22 @@ def isGConnected(deg, nodes, mal):
|
||||
|
||||
return nx.is_connected(G)
|
||||
|
||||
def getValidatorCountPerColumn(numberOfCols, numOfValidators, chiC):
|
||||
validatorCountPerColumn = dict()
|
||||
for _ in range(numOfValidators):
|
||||
colsSelected = random.sample(list(range(1, numberOfCols + 1)), chiC)
|
||||
def getNodeCountPerColumn(config):
|
||||
numberOfCols = config['numberOfColumns']
|
||||
numOfNodes = config['numberOfNodes']
|
||||
chiC1 = config['custody'] * config['validatorPerNode1']
|
||||
chiC2 = config['custody'] * config['validatorPerNode2']
|
||||
node1Count = int(numOfNodes * config['class1ratio'])
|
||||
nodeCountPerColumn = dict()
|
||||
for _ in range(numOfNodes):
|
||||
colsSelected = random.sample(list(range(1, numberOfCols + 1)), chiC1 if _ < node1Count else chiC2)
|
||||
for col in colsSelected:
|
||||
if col in validatorCountPerColumn.keys():
|
||||
validatorCountPerColumn[col] += 1
|
||||
if col in nodeCountPerColumn.keys():
|
||||
nodeCountPerColumn[col] += 1
|
||||
else:
|
||||
validatorCountPerColumn[col] = 0
|
||||
nodeCountPerColumn[col] = 0
|
||||
|
||||
return validatorCountPerColumn
|
||||
return nodeCountPerColumn
|
||||
|
||||
def runOnce(deg, validatorCountPerCol, malNodesPercentage):
|
||||
isParted = False
|
||||
@ -52,23 +57,23 @@ def runOnce(deg, validatorCountPerCol, malNodesPercentage):
|
||||
|
||||
return isPartedCount, partCount
|
||||
|
||||
def study():
|
||||
def study(config):
|
||||
degPartPercentages = dict()
|
||||
degAvgDisconnectedCols = dict()
|
||||
|
||||
for deg in degs:
|
||||
for deg in config['degs']:
|
||||
print(f"\nNetwork Degree: {deg}")
|
||||
|
||||
partPercentages = list()
|
||||
avgDisconnectedCols = list()
|
||||
for mal in mals:
|
||||
for mal in config['mals']:
|
||||
isPartedCount = partCount = 0
|
||||
validatorCountPerColumn = getValidatorCountPerColumn(numberOfColumns, numberOfValidators, custody)
|
||||
results = Parallel(-1)(delayed(runOnce)(deg, validatorCountPerColumn, mal) for _run in range(runs))
|
||||
nodeCountPerColumn = getNodeCountPerColumn(config)
|
||||
results = Parallel(-1)(delayed(runOnce)(deg, nodeCountPerColumn, mal) for _run in range(config['runs']))
|
||||
isPartedCount = sum([res[0] for res in results])
|
||||
partCount = sum([res[1] for res in results])
|
||||
partPercentages.append(isPartedCount * 100 / runs)
|
||||
avgDisconnectedCols.append(partCount / runs)
|
||||
partPercentages.append(isPartedCount * 100 / config['runs'])
|
||||
avgDisconnectedCols.append(partCount / config['runs'])
|
||||
print(f"Malicious Nodes: {mal}%, Partition Percentage: {partPercentages[-1]}, Avg. Partitions: {avgDisconnectedCols[-1]}")
|
||||
|
||||
degPartPercentages[deg] = partPercentages
|
||||
@ -80,7 +85,7 @@ def study():
|
||||
if not os.path.exists(newpath): os.makedirs(newpath)
|
||||
|
||||
conf1 = {
|
||||
'x': mals,
|
||||
'x': config['mals'],
|
||||
'y': degPartPercentages,
|
||||
'label': "NW Deg",
|
||||
'xlabel': "Malicious Node (%)",
|
||||
@ -90,7 +95,7 @@ def study():
|
||||
}
|
||||
|
||||
conf2 = {
|
||||
'x': mals,
|
||||
'x': config['mals'],
|
||||
'y': degAvgDisconnectedCols,
|
||||
'label': "NW Deg",
|
||||
'xlabel': "Malicious Node (%)",
|
||||
@ -104,13 +109,17 @@ def study():
|
||||
|
||||
|
||||
# Configuration
|
||||
runs = 10
|
||||
degs = range(6, 13, 2)
|
||||
mals = range(5, 100, 5)
|
||||
numberOfColumns = 128
|
||||
custody = 4
|
||||
numberOfNodes = 5000
|
||||
numberOfValidators = int(numberOfNodes * 2.4) # (0.8 * 1 + 0.2 * 8 = 2.4)
|
||||
config = {
|
||||
'runs': 50,
|
||||
'degs': range(6, 15, 2),
|
||||
'mals': range(5, 100, 5),
|
||||
'numberOfColumns': 128,
|
||||
'custody': 4,
|
||||
'class1ratio': 0.8,
|
||||
'validatorPerNode1': 1,
|
||||
'validatorPerNode2': 8,
|
||||
'numberOfNodes': 5000
|
||||
}
|
||||
|
||||
if __name__ == "__main__":
|
||||
study()
|
||||
study(config)
|
||||
@ -27,17 +27,21 @@ def isGConnected(deg, nodes, mal):
|
||||
|
||||
return nx.is_connected(G)
|
||||
|
||||
def getValidatorCountPerColumn(numberOfCols, numOfValidators, chiC):
|
||||
validatorCountPerColumn = dict()
|
||||
for _ in range(numOfValidators):
|
||||
colsSelected = random.sample(list(range(1, numberOfCols + 1)), chiC)
|
||||
def getNodeCountPerColumn(config, numOfNodes):
|
||||
numberOfCols = config['numberOfColumns']
|
||||
chiC1 = config['custody'] * config['validatorPerNode1']
|
||||
chiC2 = config['custody'] * config['validatorPerNode2']
|
||||
node1Count = int(numOfNodes * config['class1ratio'])
|
||||
nodeCountPerColumn = dict()
|
||||
for _ in range(numOfNodes):
|
||||
colsSelected = random.sample(list(range(1, numberOfCols + 1)), chiC1 if _ < node1Count else chiC2)
|
||||
for col in colsSelected:
|
||||
if col in validatorCountPerColumn.keys():
|
||||
validatorCountPerColumn[col] += 1
|
||||
if col in nodeCountPerColumn.keys():
|
||||
nodeCountPerColumn[col] += 1
|
||||
else:
|
||||
validatorCountPerColumn[col] = 0
|
||||
nodeCountPerColumn[col] = 0
|
||||
|
||||
return validatorCountPerColumn
|
||||
return nodeCountPerColumn
|
||||
|
||||
def runOnce(deg, validatorCountPerCol, malNodesPercentage):
|
||||
isParted = False
|
||||
@ -53,23 +57,23 @@ def runOnce(deg, validatorCountPerCol, malNodesPercentage):
|
||||
|
||||
return isPartedCount, partCount
|
||||
|
||||
def study():
|
||||
def study(config):
|
||||
nnPartPercentages = dict()
|
||||
nnAvgDisconnectedCols = dict()
|
||||
|
||||
for nn, nv in zip(numberOfNodes, numberOfValidators):
|
||||
for nn in config['numberOfNodes']:
|
||||
print(f"\nNumber of Nodes: {nn}")
|
||||
|
||||
partPercentages = list()
|
||||
avgDisconnectedCols = list()
|
||||
for mal in mals:
|
||||
for mal in config['mals']:
|
||||
isPartedCount = partCount = 0
|
||||
validatorCountPerColumn = getValidatorCountPerColumn(numberOfColumns, nv, custody)
|
||||
results = Parallel(-1)(delayed(runOnce)(deg, validatorCountPerColumn, mal) for _run in range(runs))
|
||||
nodeCountPerColumn = getNodeCountPerColumn(config, nn)
|
||||
results = Parallel(-1)(delayed(runOnce)(config['deg'], nodeCountPerColumn, mal) for _run in range(config['runs']))
|
||||
isPartedCount = sum([res[0] for res in results])
|
||||
partCount = sum([res[1] for res in results])
|
||||
partPercentages.append(isPartedCount * 100 / runs)
|
||||
avgDisconnectedCols.append(partCount / runs)
|
||||
partPercentages.append(isPartedCount * 100 / config['runs'])
|
||||
avgDisconnectedCols.append(partCount / config['runs'])
|
||||
print(f"Malicious Nodes: {mal}%, Partition Percentage: {partPercentages[-1]}, Avg. Partitions: {avgDisconnectedCols[-1]}")
|
||||
|
||||
nnPartPercentages[nn] = partPercentages
|
||||
@ -81,7 +85,7 @@ def study():
|
||||
if not os.path.exists(newpath): os.makedirs(newpath)
|
||||
|
||||
conf1 = {
|
||||
'x': mals,
|
||||
'x': config['mals'],
|
||||
'y': nnPartPercentages,
|
||||
'label': "Nodes",
|
||||
'xlabel': "Malicious Node (%)",
|
||||
@ -91,7 +95,7 @@ def study():
|
||||
}
|
||||
|
||||
conf2 = {
|
||||
'x': mals,
|
||||
'x': config['mals'],
|
||||
'y': nnAvgDisconnectedCols,
|
||||
'label': "Nodes",
|
||||
'xlabel': "Malicious Node (%)",
|
||||
@ -105,13 +109,17 @@ def study():
|
||||
|
||||
|
||||
# Configuration
|
||||
runs = 10
|
||||
deg = 8
|
||||
mals = range(5, 100, 5)
|
||||
numberOfColumns = 128
|
||||
custody = 4
|
||||
numberOfNodes = [int(_) for _ in (np.logspace(2, 4, 5, endpoint=True, base=10) * 5)]
|
||||
numberOfValidators = [int(nn * 2.4) for nn in numberOfNodes] # (0.8 * 1 + 0.2 * 8 = 2.4)
|
||||
config = {
|
||||
'runs': 10,
|
||||
'deg': 8,
|
||||
'mals': range(5, 100, 5),
|
||||
'numberOfColumns': 128,
|
||||
'custody': 4,
|
||||
'class1ratio': 0.8,
|
||||
'validatorPerNode1': 1,
|
||||
'validatorPerNode2': 8,
|
||||
'numberOfNodes': [int(_) for _ in (np.logspace(2, 4, 5, endpoint=True, base=10) * 5)]
|
||||
}
|
||||
|
||||
if __name__ == "__main__":
|
||||
study()
|
||||
study(config)
|
||||
Loading…
x
Reference in New Issue
Block a user