The native library compression SDK, the author developed, can help app developer to integrate the LZMA native library compression to obtain smaller files.
Besides the LZMA compression, this SDK provides additional functions to encourage developers to include an x86 native library with the following:
LZMA significantly improves compression ratio in comparison to LZ77-based algorithms at the cost of much slower compression (decompression speed is not much affected).
We found experimentally that a transform optimized for LZMA and BWT-based  (e.g.
PPMVC offers compression ratio higher than LZMA, and taster compression time.
Our implementation of SDHT and SHT has embedded four back-end compression algorithms: gzip, LZMA, PPMVC, and FastPAQ8.
For comparison purposes, we included in the tests general-purpose compression tools: gzip 1.2.4, LZMA 4.43, PPMVC 1.2, and FastPAQ8, employing the same algorithms at the final stage of SDHT and SHT, to demonstrate the improvement from applying the HTML transform.
The next parts of Table 2 contain compression results of the introduced HTML corpus using gzip, LZMA, PPMVC, FastPAQ, and our implementation of the SDHT and SHT algorithms combined with gzip, LZMA, PPMVC, and FastPAQ.
Compared to the general-purpose compression tools, SDHT improves compression of the introduced HTML corpus on average by 4% in case of gzip, 0% for LZMA, almost 2% in case of PPMVC and about 1% for FastPAQ.