Add documentation for JLD2 I/O

docs/src/man/importing_and_exporting.md

@@ -113,6 +113,56 @@ As you can see the code required to transform `iris` into a proper input to the
 format is not easy. Therefore CSV.jl is the preferred package to write CSV files
 for data stored in data frames.
 
+## JLD2 Files
+
+JLD2 is a HDF5-compatible file format that allows reading and writing data frames.
+A valuable feature of JLD2 format is that it preserves custom column types of the stored data frame.
+
+The `save` and `load` functions, provided by FileIO.jl, allow to read/write a data frame from/to a JLD2 file.  
+
+A data frame can be saved as a JLD2 file output.jld2 using
+
+If you have not used the FileIO and JLD2 packages before then you may need to install it first:
+```julia
+using Pkg; 
+Pkg.add("FileIO")
+Pkg.add("JLD2")
+```
+
+The FileIO and JLD2 functions are not loaded automatically and must be imported into the session.
+```julia
+using FileIO
+using JLD2
+```
+
+We can now create a simple data frame and save it as a jld2 file using `save`. `save` 
+accepts an AbstractDict yielding the key/value pairs, where the key is a string representing 
+the name of the dataset and the value represents its contents:
+
+```julia
+df = DataFrame(x=1, y=2)
+save("output.jld2", Dict("df" => df))
+```
+
+A jld2 file can be read in using `load`. If `load` is called with a single dataset name, 
+load returns the contents of that dataset from the file:
+```julia
+df = load("output.jld2", "df")
+1×2 DataFrame
+ Row │ x      y     
+     │ Int64  Int64 
+─────┼──────────────
+   1 │     1      2
+```
+
+If `load` is called with a filename argument only, the load function loads all datasets 
+from the given file into a Dict:
+```julia
+df = load("output.jld2")
+Dict{String, Any} with 1 entry:
+  "df" => 1×2 DataFrame…
+```
+
 ## Other formats
 
 Other data formats are supported for reading and writing in the following packages