tables.toml

[top_comments]
Overview = """
These benchmarks compare the performance of the frozen collecitons relative
to the classic Rust collections.

The frozen collections have different optimizations depending on the type of data they
store and how it is declared. The benchmarks probe those different features to show
the effect of the different optimizations on effective performance.

When you see `HashSet(classic)` vs. `HashSet(foldhash)` this reflects the performance difference between the
normal hasher used by the standard collections as opposed to the performance that the
`foldhash` hasher provides.

The benchmarks assume a 50% hit rate when probing for lookup, meaning that
half the queries are for non-existing data. Some algorithms perform differently between
present vs. non-existing cases, so real world performance of these algorithms depends on the
real world hit rate you experience.
"""

[table_comments]

dense_scalar = """
Scalar sets where the values are in a contiguous range.
"""

sparse_scalar = """
Scalar sets where the values are in a non-contiguous range.
"""

random_scalar = """
Scalar sets where the values are randomly distributed.
"""

random_string = """
String sets where the values are random.
"""

prefixed_string = """
String sets where the values are random, but share a common prefix.
"""

hashed = """
Sets with a complex key type that is hashable.
"""

ordered = """
Sets with a complex key type that is ordered.
"""