Proxies for web scraping: a fixed pool of IPs with unmetered traffic
Updated:
Web scraping proxies billed per gigabyte make a crawler more expensive every time a page gets heavier, which is why steady, scheduled crawlers are cheaper on a fixed pool of addresses with unmetered traffic. ProxyV4 sells private IPv4 in seven countries and static ISP in the US and Vietnam, each priced per IP with no traffic metering. You size the pool by how many requests per second the targets tolerate, and the bill stays the same whatever the pages weigh.
Per-IP pricing versus per-GB pricing for a crawler
| Question | Static pool (per IP) | Metered traffic (per GB) |
|---|---|---|
| What drives the cost | Number of IPs and rental term | Bytes downloaded, including HTML, JS and assets |
| What happens when pages get heavier | Nothing; bandwidth is unlimited | The bill grows with them |
| What limits throughput | Each target's per-IP rate limit | Budget |
| How to scale | Add IPs; the price per IP drops with quantity | Buy more traffic |
Why scrapers get blocked, and what a pool changes
On catalogue, listings and public API scrapes, most blocks are not about who you are but how fast you ask: one address exceeding a site's request rate gets HTTP 429. A pool of 10 to 100 static IPs, each kept at a polite concurrency, keeps every address under the threshold while the crawler as a whole runs at the speed you need.
Single tenancy helps here too. Each ProxyV4 IP belongs to one customer for the whole rental, so the rate history a target sees on that address is yours alone.
Matching the pool to the target
- Public catalogues and high-volume APIs: private IPv4 in the United States or Singapore, the lowest cost per IP
- Vietnamese e-commerce and local portals: Vietnam IPv4 to get domestic pages and local prices
- Targets that filter hosting ASNs behind a WAF: static ISP in the United States or Vietnam, registered to consumer ISPs, which tends to mean fewer Cloudflare challenges
- Targets that need a new IP on every request: rotating Vietnam proxies are not on sale yet; the rotating Vietnam page has the waitlist
Crawler settings that make a static pool last
Every line uses standard credentials, so in Python the proxy is just proxies = {"http": "http://user:pass@ip:port", "https": "http://user:pass@ip:port"}.
- Assign requests round-robin across the list and add random jitter, roughly 2 to 5 seconds per IP per target domain
- Reuse connections with
requests.Session()or a pooled client so you are not paying for a TLS handshake on every request - In Playwright and Puppeteer, block images, fonts and media you do not parse
- When an IP gets HTTP 429 or 503, take it out of rotation for a cooldown and back off instead of retrying immediately
Step by step
- 1
- 2
Turn the list into URLs
Paste the
ip:port:user:passlines into the proxy format converter to gethttp://user:pass@ip:portstrings or JSON. - 3
Plug the pool into your framework
Use the setup pages for Python Requests, Scrapy, Selenium or Node.js Axios.
- 4
Measure before the first full run
Check response times with the proxy speed test and drop any line that is slow to your target.
Frequently asked questions
Is scraping traffic on ProxyV4 proxies metered?
No. Private IPv4 and static ISP are priced per IP for the rental term with unlimited bandwidth.
How many pages a day can one static IP fetch?
It depends on the target's rate limit. At 1 to 3 requests per minute per domain, one IP fetches roughly 1,500 to 4,000 pages a day without standing out.
Do you sell rotating proxies for scraping?
Not yet. Rotating Vietnam proxies are waitlist only. For now, a static pool with round-robin assignment does the same job for most scheduled crawls.
How do I convert 50 proxies into Scrapy format quickly?
Paste them into the proxy format converter. It turns ip:port:user:pass lines into URL strings or a JSON array in the browser.
Build a scraping pool with a fixed monthly cost
Private IPv4 in seven countries and static ISP in the US and Vietnam, priced per IP with unlimited traffic.
Check live US IPv4 prices