Dask提供了许多不需要使用Dashboard的监测内存使用情况的方法。以下是几种方法的示例：

1. 使用Client.profile()方法

from dask.distributed import Client

client = Client()

# Collect profile data for a few seconds
prof_data = client.profile(start=0, duration=5)

# Print memory usage in bytes
print(prof_data["worker"]["memory"] / 1024**2)

2. 使用Client.resource_monitor()方法

from dask.distributed import Client

client = Client()

# Enable resource monitoring on all workers
client.resource_monitor(duration="10s")

# Access the collected resource information
for host, data in client.resources().items():
    print(f"Host: {host} | Memory (GB): {data['memory']['total'] / 1e9}")

3. 使用Client.run()方法

from dask.distributed import Client

client = Client()

# Define function to check memory usage of each worker
def check_mem():
    import psutil
    mem = psutil.virtual_memory()
    return mem.available / 1024**3

# Check memory usage on each worker
mem_usage = client.run(check_mem)

# Print results
for worker, mem in mem_usage.items():
    print(f"Worker {worker}: {mem:.2f} GB")

以上方法不需要使用Dask的Dashboard即可检查Dask的内存使用情况。