(algorithm)

**Definition:**
*Merge* n sorted sequences of different lengths into one output while minimizing reads. Only two sequences can be merged at once. At each step, the two shortest sequences are merged.

**Formal Definition:** Let D={n_{1}, ... , n_{k}} be the set of lengths of sequences to be merged. Take the two shortest sequences, n_{i}, n_{j}∈ D, such that n≥ n_{i} and n≥ n_{j} ∀ n∈ D. Merge these two sequences. The new set D is D' = (D - {n_{i}, n_{j}}) ∪ {n_{i}+n_{j}}. Repeat until there is only one sequence.

**See also**
*simple merge*, *ideal merge*, *Huffman coding*, *greedy algorithm*.

*Note:
Merging sequences by length is the same as joining trees by frequency in Huffman coding. For example, let there be a set of sorted sequences of the following lengths: D={3,5,7,9,12,14,15,17}. Building the optimal merge tree goes as follows. Note that merged sequences are replaced by the sum of their lengths. For instance, the first step merges the sequence of length 3 and the sequence of length 5 to get a sequence of length 8.
*

3 5 7 9 12 14 15 17

8 7 9 12 14 15 17

/ \

3 5

15 9 12 14 15 17

/ \

8 7

/ \

3 5

15 21 14 15 17

/ \ / \

8 7 9 12

/ \

3 5

29 21 15 17

/ \ / \

14 15 9 12

/ \

8 7

/ \

3 5

29 21 32

/ \ / \ / \

14 15 9 12 15 17

/ \

8 7

/ \

3 5

50 32

/ \ / \

/ \ 15 17

29 21

/ \ / \

14 15 9 12

/ \

8 7

/ \

3 5

82

/ \

/ \

/ \

50 32

/ \ / \

/ \ 15 17

29 21

/ \ / \

14 15 9 12

/ \

8 7

/ \

3 5

Author: AL

Go to the Dictionary of Algorithms and Data Structures home page.

If you have suggestions, corrections, or comments, please get in touch with Paul Black.

Entry modified 14 August 2008.

HTML page formatted Mon Feb 2 13:10:40 2015.

Cite this as:

Alen Lovrencic, "optimal merge", in
*Dictionary of Algorithms and Data Structures* [online], Vreda Pieterse and Paul E. Black, eds. 14 August 2008. (accessed TODAY)
Available from: http://www.nist.gov/dads/HTML/optimalMerge.html