The nature of substitutions embedded in JSON string values and YAML scalar nodes means in large blueprints that use a lot of substitutions, many embedded strings need to be parsed individually.
When running on a server with a large amount of blueprints being parsed simultaneously, this can become a serious performance concern.
One solution would be to cache the blueprint schema with substitutions expanded in-memory or in a serialised format in a database or on a file system based on the needs of the system running the blueprint framework. (in-memory would not be a good solution for a context where horizontal scalability is important)
The nature of substitutions embedded in JSON string values and YAML scalar nodes means in large blueprints that use a lot of substitutions, many embedded strings need to be parsed individually. When running on a server with a large amount of blueprints being parsed simultaneously, this can become a serious performance concern. One solution would be to cache the blueprint schema with substitutions expanded in-memory or in a serialised format in a database or on a file system based on the needs of the system running the blueprint framework. (in-memory would not be a good solution for a context where horizontal scalability is important)