I don't know what you mean by "huge". Huge is a relative word and we each have our definition for it.
Anyway, assuming it is not that big, you can use the System.Xml.XmlDocument class in the .NET Framework 2.0. It gives you a great deal of flexibility and control over your XML document.
If you are just looking to parse the data and if the file is indeed "huge", then I suggest you use the System.Xml.XmlReader class and store the throughputs to a System.Collections.Generic.Dictionary<T,T> class. The dictionary class is easily the fastest collection type if you are not bothered about sorting. It can manage literally manage millions of records quickly.