libmmmulti 0.1-2 source package in Ubuntu

 Sometimes you have a lot of plain-old data, but you need random access
 to it. These header-only classes combine memory-mapped files with
 high-performance parallel sorting and appropriate indexing strategies
 to support very large (>memory but <disk) multimaps, multisets, and
 interval trees.
 .
 This implements a memory backed multimap intended for use where:
  * your keys are integers, or can be mapped to dense range of integers,
  * the memory mapped file is on fast storage, like an SSD (although
    this is not a requirement),
  * you have arbitrary values of fixed size (e.g. structs, other POD
    types) that can be sorted,
  * you don't need dynamic updates of the table,
  * and you are likely to run out of memory of you use a traditional
    map or hash table,
  * but you can handle approximately 1 bit per record in RAM.
 .
 These may seem to be very specific, but many problems can be mapped into
 a dense integer set. mmmulti::map developed first as a data structure
 to support seqwish, which uses it to generate precise variation graphs
 from pairwise alignments between collections of sequences. As this
 multimap forms a key data processing kernel in the algorithm, it can
 scale to extremely large problem sizes, limited only by available disk
 space. Although performance is much slower than an in-memory structure,
 it is virtually guaranteed to be able to complete the compute.

 This implements a memory backed multimap intended for use where:
  * your keys are integers, or can be mapped to dense range of integers,
  * the memory mapped file is on fast storage, like an SSD (although
    this is not a requirement),
  * you have arbitrary values of fixed size (e.g. structs, other POD
    types) that can be sorted,
  * you don't need dynamic updates of the table,
  * and you are likely to run out of memory of you use a traditional
    map or hash table,
  * but you can handle approximately 1 bit per record in RAM.
 .
 These may seem to be very specific, but many problems can be mapped into
 a dense integer set. mmmulti::map developed first as a data structure
 to support seqwish, which uses it to generate precise variation graphs
 from pairwise alignments between collections of sequences. As this
 multimap forms a key data processing kernel in the algorithm, it can
 scale to extremely large problem sizes, limited only by available disk
 space. Although performance is much slower than an in-memory structure,
 it is virtually guaranteed to be able to complete the compute.