Choose from six locations: Singapore, Tokyo, Seoul, Hong Kong, the US East and the US West. Each order provides a dedicated physical machine, not a virtual machine. Listed configurations are generally available; check the console for live availability.
Distance from developers to a location is only one factor. Code repositories, dependency sources, artifact storage and teammates all affect real-world build times.
For frequent SSH or VNC sessions, compare sustained median latency and jitter from your network to candidate locations. Don't decide based on a single best-case result.
Repository clones, submodule syncs and large file downloads are part of the pipeline. A location close to developers but far from the repository can still mean a slow initial checkout.
Check access routes to CocoaPods, Swift Package Manager, npm, Flutter and internal artifact repositories. Record results separately with cold and warm caches.
For CI, consider who monitors queues, handles failures and retrieves artifacts. Distributed teams may benefit from placing the machine near the primary on-call team or main artifact repository.
Use the same repository revision, dependency cache state and Xcode settings. Track checkout, dependency resolution, Archive and artifact upload times separately before choosing a long-term location.
All four locations offer CDNMac M4: M4, 16GB RAM and 256GB SSD. Location does not change the base machine specs. Confirm optional add-ons separately when ordering.
A good fit for Southeast Asia-based developers, regional repositories and Asia-Pacific dependency sources. If your team is spread across regions, also test repository pulls and artifact uploads.
A good fit for graphics sessions, code checkouts and CI jobs in Japan and Northeast Asia. If your artifact repository is in another region, track upload times too.
A good fit for developers in South Korea and workflows across Northeast Asia. For frequent remote Xcode GUI sessions, pay close attention to sustained latency and jitter.
A good fit for cross-border collaboration in East Asia, code sync and remote administration. If dependencies are spread across regions, measure cold-cache downloads and artifact uploads separately.
The US directory includes two locations: US East and US West. We don't split US West into unlisted cities or claim additional regions.
A good fit for workflows with key team members, code repositories or artifact storage in eastern North America. If Asia-Pacific teammates need frequent GUI access, test the sustained interactive experience first.
A good fit for teams in western North America and build pipelines that also need Asia-Pacific connectivity. Test access to your repository, dependency sources and artifact storage before choosing.
One model is currently offered. The matrix shows which models are listed at each location; it doesn't indicate real-time machine assignments. Check the console for live availability.
| Model | Singapore | Tokyo, Japan | Seoul, South Korea | Hong Kong | US East | US West |
|---|---|---|---|---|---|---|
| CDNMac M4 M4 · 16GB · 256GB | Available | Available | Available | Available | Available | Available |
The results below compare network round-trip latency; they don't promise build times. Tests used fixed broadband and public IPv4 with a consistent sampling method. All figures are in milliseconds.
| Test location and ISP | Singapore | Tokyo | Seoul | Hong Kong | US East | US West |
|---|---|---|---|---|---|---|
| SingaporeSingtel fixed broadband | 4 ms | 68 ms | 79 ms | 34 ms | 230 ms | 175 ms |
| TokyoNTT fixed broadband | 69 ms | 4 ms | 35 ms | 49 ms | 170 ms | 112 ms |
| SeoulKT fixed broadband | 77 ms | 34 ms | 4 ms | 45 ms | 190 ms | 132 ms |
| Hong KongHKT fixed broadband | 35 ms | 47 ms | 42 ms | 3 ms | 218 ms | 159 ms |
| US East test pointVerizon Fios | 232 ms | 169 ms | 188 ms | 215 ms | 8 ms | 72 ms |
| US West test pointAT&T Fiber | 176 ms | 111 ms | 131 ms | 158 ms | 73 ms | 7 ms |
For each retest, send 3 warm-up requests, then collect 30 valid samples and use the median. Public internet routing varies by ISP, connection type and time of day, so test repeatedly from your own network before choosing a location.
Low round-trip latency usually helps interactive sessions, but it represents only one part of the network path. Builds, downloads and graphical sessions depend on other factors too.
The ICMP median shows the basic round-trip path from a test point to a location. It doesn't measure TCP throughput, packet loss recovery, inter-network congestion or long-lived connection stability.
Dependency source response times, cache hits, repository size, build targets and signing steps all affect total time. Track each stage rather than just the full pipeline duration.
VNC performance also depends on jitter, packet loss, how often the screen changes and local display resolution. For extended GUI work, test editing, archiving and log review in practice.
First measure sustained network latency. Then clone a real repository, restore dependencies, run an Archive, and verify artifact uploads and remote administration. These four results are more useful for choosing a location than a single ping.
Singapore, Tokyo, Seoul, Hong Kong, US East and US West are all in the current directory. Check the console for live stock, delivery times and available add-ons.