Start by choosing a region, then compare connection types. Representative cities can help narrow your options; check the client for currently available servers.
The table lists common data center cities and connection types worth comparing first; it is not a live server list. Use cases suggest what to check when testing a route, not compatibility or performance guarantees. If a city or route type isn’t listed in your client, compare the options actually available.
Asia-Pacific
Country/Region
Representative city
Route type
Best for
Hong Kong, China
Hong Kong
IEPL private line
Video calls and remote work
Japan
Tokyo
Direct
Everyday browsing and games in Japan
Japan
Osaka
Relay
Backup route for a Japan exit
Singapore
Singapore
Relay
Southeast Asian business and cloud documents
South Korea
Seoul
Direct
South Korean websites and games
Australia
Sydney
Relay
Accessing Australian websites
North America
Country/Region
Representative city
Route type
Best for
United States
Los Angeles
Relay
West Coast websites and streaming responses
United States
Seattle
Direct
Accessing West Coast cloud services
United States
New York
Relay
Accessing East Coast business systems
Canada
Toronto
Relay
Canadian websites and remote collaboration
Canada
Vancouver
Direct
Accessing websites in western Canada
Europe
Country/Region
Representative city
Route type
Best for
Germany
Frankfurt
Relay
Accessing European cloud services
United Kingdom
London
Direct
UK websites and online meetings
France
Paris
Relay
French websites and video streaming
Netherlands
Amsterdam
Relay
Backup for Western European business
Spain
Madrid
Direct
Accessing local Spanish websites
Other regions
Country/Region
Representative city
Route type
Best for
Brazil
São Paulo
Relay
Brazilian websites and regional business
Chile
Santiago
Relay
Accessing Chilean websites
United Arab Emirates
Dubai
Relay
Middle Eastern business and cloud collaboration
South Africa
Johannesburg
Relay
Accessing South African websites
Route Types and Cost Differences
The region answers “where does the connection exit?”; the route type answers “how does it get there?” The connection method can differ even when the exit city is the same. Route types describe network paths, not rankings of video quality, speed, or service compatibility.
Dedicated transport segment
IEPL private line
IEPL is an international Ethernet private line, typically used for a segment of the connection between an entry point and a remote access point. Compared with routing the entire path over the public internet, a dedicated segment is easier to manage. It’s worth comparing its performance during your usual hours for video calls, remote desktops, and sustained uploads.
Dedicated transport resources and maintenance generally cost more, but other networks can still affect the path from your device to the entry point and from the exit to the destination website. A dedicated line may not help if your home Wi-Fi is congested. First check whether the issue is your local signal or whether changing routes actually improves the same task.
Forwarded through an access point
Relay route
A relay sends your device’s traffic to an access point first, then forwards it to an exit in the destination region. This adds a connection stage and maintenance costs, but the route from entry to exit can differ from a direct route over your local network. If a direct connection often reconnects or drops during your usual hours, try a relay while keeping the same exit region.
Whether a relay is worthwhile depends on the access point’s location, the onward route, and the time of day. The same exit doesn’t mean the same entry point, and a nearby entry point doesn’t guarantee a smooth connection for the rest of the route. When comparing options, keep the region, device, and app the same, and test with the same video or file. Don’t rule out a relay just because it adds a hop.
Direct connection to the exit
Direct route
With a direct route, the client connects to the exit server without an additional service-side access point; traffic still passes through your local ISP and the public internet. The simpler connection setup often means lower service-side relay costs. If the route from your location to the exit is smooth, try direct for everyday browsing, research, or short sessions.
Geographic distance is only one factor: routes between ISPs don’t always follow the shortest physical path. A direct connection that works well at home may perform differently on an office network. Compare routes on the networks you actually use instead of judging by the route name alone. These cost differences reflect network structure; see the plans page for subscription prices.
Choose a route by Use Case
When testing routes, change one variable at a time. Keep the device, network, and target app the same, then compare different routes in the same region. If you need to change regions, first check whether the app requires a specific exit location.
Everyday browsing: Check that pages fully load
If the service doesn’t require a specific region, start with a direct route in a nearby region. Open a news or document site you use regularly, then check pages with images or scripts as well as the homepage. Sometimes the text appears quickly while images, search, or the sign-in page takes much longer, so judging by the initial load alone can be misleading.
If only one site has a problem, try other sites on the same route first. If they work, check the target site and your browser settings. If several sites keep disconnecting at the same time, compare a relay in the same region. Test during your usual browsing hours rather than relying on results from a quieter time to predict evening performance.
Streaming: Tell content availability from buffering
Check the platform’s regional rules to confirm the content is available, then compare routes using the same title and video quality. Loading the homepage doesn’t mean a particular title will play; available content can also depend on your account region, licensing, and device. If you see a region notice, check the platform’s guidance rather than assuming you need more bandwidth.
If a video starts but keeps buffering, check what happens when you switch to higher quality, skip ahead, or let it play continuously. If buffering happens only at certain times, pause local syncing or downloads, then compare other available routes in the same region. If lowering the video quality clearly helps, sustained transfer conditions may be the issue. For more troubleshooting tips, see the streaming guide.
AI tools: Check sign-in and streaming responses
Check the tool’s published service regions and account requirements, then choose an exit that meets them. Observe what happens at each stage: opening the sign-in page, starting a session, and waiting for the full response. Sign-in failures may be due to an expired session or account permissions; if the response starts and then stalls, check persistent connections and the client’s traffic routing settings.
For a fair comparison, use the same kind of request. Confirm that short replies work before testing a longer response. If the website works but the command line or editor extension doesn’t, check whether the tool uses the same network settings. If the platform applies rate limits, follow its instructions and wait; switching regions repeatedly won’t fix account quotas or permissions. See the AI acceleration handbook.
Gaming: Test routes for your game region
First check which game region you’re matched to, then compare exits nearby. A game’s launcher, sign-in service, and live matches may use different regions, so a smooth sign-in doesn’t guarantee a stable match. In a practice area or casual mode, check for character rubber-banding, intermittent input lag, and voice chat dropouts.
If a direct route fluctuates during your usual play hours, compare it with a relay in the same region. An extra segment can add latency, but it may also avoid a poor route. Don’t switch routes mid-match; changing exits usually means reconnecting. If a wired connection works but Wi-Fi doesn’t, address local wireless interference first.
Work: Test uploads and calls
Follow your organization’s network access policies and test with real work tasks. In an authorized environment, open a cloud document, upload a non-sensitive file, then test a voice or video call. A website loading successfully won’t tell you whether uploads keep dropping or long calls stay connected. During busy work hours, compare a relay and a dedicated line in the same exit region.
If your team’s systems restrict access by region or exit address, ask your administrator which connection methods are allowed; don’t try to fix a permissions error by switching routes. Keep a note of your usual route and a tested backup. Save your work before switching, and avoid doing so during a call or upload. New to client setup? Start with the setup guide.