1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
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
58
59
|
"""
A spider that generate light requests to measure QPS throughput
usage:
scrapy runspider qpsclient.py --loglevel=INFO --set RANDOMIZE_DOWNLOAD_DELAY=0
--set CONCURRENT_REQUESTS=50 -a qps=10 -a latency=0.3
"""
from scrapy.http import Request
from scrapy.spiders import Spider
class QPSSpider(Spider):
name = "qps"
benchurl = "http://localhost:8880/"
# Max concurrency is limited by global CONCURRENT_REQUESTS setting
max_concurrent_requests = 8
# Requests per second goal
qps = None # same as: 1 / download_delay
download_delay = None
# time in seconds to delay server responses
latency = None
# number of slots to create
slots = 1
def __init__(self, *a, **kw):
super().__init__(*a, **kw)
if self.qps is not None:
self.qps = float(self.qps)
self.download_delay = 1 / self.qps
elif self.download_delay is not None:
self.download_delay = float(self.download_delay)
async def start(self):
for item_or_request in self.start_requests():
yield item_or_request
def start_requests(self):
url = self.benchurl
if self.latency is not None:
url += f"?latency={self.latency}"
slots = int(self.slots)
if slots > 1:
urls = [url.replace("localhost", f"127.0.0.{x + 1}") for x in range(slots)]
else:
urls = [url]
idx = 0
while True:
url = urls[idx % len(urls)]
yield Request(url, dont_filter=True)
idx += 1
def parse(self, response):
pass
|
