1. What is the lifetime of the external data compared to the lifetime of any of the accessing Lua VM?
Each lua vm may access any part of the dataset. The dataset is immutable, but can be update to a new version. When the dataset is updated, all the following accessing turn to the new version, but old version may be used a while. Because many coroutines in each vm, some coroutine may begin before updating and finish after a while after updating.
2. What entity interprets the external data (Lua or C code) to impute it’s structure and any deducible descriptive metadata?
One lua vm interprets the datas, and update it to a new version. It send message to others when the external dataset updated. The dataset can be schema-less (like mongodb's document)
3. How does the data need to be presented to Lua for processing (numbers, strings, opaque data etc)?
The data can be numbers, strings, and tables. (Table can be serialized to a message)
4. What are the operations that Lua needs to perform on this data, and what tradeoffs are there between size and speed for these operations?
We need read the numbers and strings, and iterate the table for serialization or as arguments pass to some functions.
Size is important than speed.
