Dedicated physical Mac mini nodes are available in Singapore, Tokyo, Seoul, Hong Kong, and the eastern US. Both configurations are generally orderable across all five locations; live availability returned by the console at checkout is authoritative.
First filter by your team’s and repositories’ actual network paths, then compare latency. “Available” on a node card means the configuration and region are open in the sales directory; it is not a static inventory snapshot.
For Southeast Asian teams, regional build pipelines, and workloads that need proximity to repositories or artifact storage in Singapore.
Ideal for development teams in Japan and East Asia, as well as workloads running Xcode builds and remote collaboration in the Tokyo time zone.
Designed for Korea and Northeast Asia access paths, and for Seoul-based teams running continuous builds and remote development environments.
A strong fit for collaboration between South China and Southeast Asia, cross-region builds, artifact delivery, and remote macOS development.
For collaboration across Eastern North America and Western Europe, especially when repositories, team members, or delivery targets span the Atlantic.
Berry M4 and Berry M4 Pro are dedicated physical Mac mini nodes, not virtual machines. All directory combinations are open; check the live availability returned by the console before placing an order.
| Available configurations | Singapore | Japan (Tokyo) | South Korea (Seoul) | Hong Kong | Eastern US |
|---|---|---|---|---|---|
| Berry M4 M4 · 16GB · 256GB | Available | Available | Available | Available | Available |
| Berry M4 Pro M4 Pro · 64GB · 2TB | Available | Available | Available | Available | Available |
Every BerryVM directory combination can start an order; the console returns final availability live while configuring the order.
The table records ICMP round-trip times from each test city to the five nodes. Each path was sampled 30 times; timeout samples were removed before calculating the median. Use these figures for initial screening—actual connections vary by local carrier, routing, and time of day.
| Test origin city | Singapore | Japan (Tokyo) | South Korea (Seoul) | Hong Kong | Eastern US |
|---|---|---|---|---|---|
| Shanghai | 71 ms | 49 ms | 44 ms | 31 ms | 204 ms |
| Taipei | 53 ms | 33 ms | 42 ms | 18 ms | 176 ms |
| Bangkok | 32 ms | 87 ms | 82 ms | 47 ms | 227 ms |
| Frankfurt | 164 ms | 238 ms | 245 ms | 201 ms | 92 ms |
| New York | 242 ms | 171 ms | 185 ms | 221 ms | 18 ms |
Prioritize median latency and variation from team members to the node. Remote desktops, code browsing, and debugging are more sensitive to path stability than to a single best-case result.
When developers do not operate a graphical interface directly, include the paths to repositories, caches, and artifact destinations—not just the office location.
For shared pipelines, choose the node with the most balanced overall paths, then retest after activation using real repositories and artifact transfers.
Choosing a node is not simply about picking the city closest to one person. For build pipelines, repositories, dependency caches, artifact storage, and collaboration time zones may matter more than the operator’s location.
List the main cities of daily users, then prioritize round-trip latency, packet loss, and peak-hour variation for remote sessions.
Builds that frequently pull source code and dependencies should measure the complete path from the node to repositories, package mirrors, and cache services.
If artifacts must be uploaded to a specific region, compare real file-transfer times rather than substituting ICMP figures for throughput testing.
When people need to take over manually, inspect logs, or handle support tickets, the node’s time zone affects handoffs and troubleshooting efficiency.
Record the test city, network access, protocol, sample count, and median. Retest under the same conditions after activation to identify whether changes come from the local line, cross-border routing, or the workload itself.
After you choose an entry point, the configuration sheet preselects the corresponding node. Then confirm Berry M4 or Berry M4 Pro, daily, weekly, monthly, or quarterly billing, plus storage and Thunderbolt 5 parallel-connection options.
This page shows the long-term sales directory. When you submit an order, the console returns the current availability of the selected node and configuration. Verify the node, term, additional storage, and parallel-connection count before payment.
Changing the node of a dedicated physical Mac involves device reassignment and confirmation of data handling. Submit a support ticket through the console first; do not send access passwords or unredacted logs in a ticket.
Berry M4 starts at $20.4/day. All five locations support Berry M4 and Berry M4 Pro, with daily, weekly, monthly, and quarterly terms. Billing is in USD; payment supports only USDT-TRC20 and Visa / Mastercard / Amex (via Stripe).