greg/pcompress
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
160 commits 3 branches 21 tags 7.6 MiB
Commit graph

68 commits

Author SHA1 Message Date
Moinak Ghosh
d9eb82e0e8 Fix numeric parsing. 
Fix dedupe bug introduced in last commit.
Reset valid flag when resetting dedupe context.
Cleanup test suites.
Do not abort test suite on failure of a test case.
 2013-01-03 00:27:18 +05:30
Moinak Ghosh
13d9378acd Update to latest XXHash version. 2012-12-31 11:53:47 +05:30
Moinak Ghosh
28224d29d3 Improve Dedupe performance. 
Add more debug timing stats.
Change default checksum to Keccak 256 (SIMD version 4x faster than Skein).
Fix compiler warning in allocator code.
 2012-12-29 23:43:41 +05:30
Moinak Ghosh
26a4f42506 Introduce strict compiler flags and fix scores of warnings/issues. 
Avoid different optimization flags for Dedupe sources.
Fix liberal mixing of uint64_t and int64_t (should all be uint64_t).
Fix corner case crash when decompressing.
 2012-12-27 23:06:48 +05:30
Moinak Ghosh
a43fdd7d2c Improve Delta2 scanning speed and effectiveness. 
Add destination buffer overflow check in Delta2.
Add rough speed computation.
 2012-12-23 00:44:56 +05:30
Moinak Ghosh
224fb529e9 Get rid of size_t in places where 64-bitness is assumed. 2012-12-09 10:15:06 +05:30
Moinak Ghosh
6c3173f929 Work in progress global dedupe config setup. 2012-11-29 22:28:50 +05:30
Moinak Ghosh
d250322490 Fix issues with error handling. 
Add new tests for out of range values and corrupted file.
 2012-11-24 23:53:07 +05:30
Moinak Ghosh
d054e0f713 Zlib optimizations. Use raw deflate streams to avoid unnecessary adler32. 
Change some function signatures to improve algo init function behavior.
Fix corner case dedupe bug in error handling flow.
Bump archive version signature.
 2012-11-22 21:02:50 +05:30
Moinak Ghosh
3b1d6b55fe Work in progress global dedupe config loader. 2012-11-19 21:41:56 +05:30
Moinak Ghosh
393ced991a A couple of minor cleanups. 2012-11-18 20:20:16 +05:30
Moinak Ghosh
b2cbf0699e Use fixed rolling-hash mask for better block size approximation. 2012-11-11 14:58:39 +05:30
Moinak Ghosh
eacbf207aa Tweak chunking parameters for better block size distribution and dedupe ratio. 2012-11-08 19:41:33 +05:30
Moinak Ghosh
e437390e53 Add some more debug mode info. 2012-11-07 22:55:14 +05:30
Moinak Ghosh
c5ebe1f30a Portability to Debian based distros. 
Enable SSE4/AVX detection for AMD platforms (Bulldozer has both).
Portable long long int print formatting to silence gcc 4.6 warnings.
 2012-10-21 21:03:07 +05:30
Moinak Ghosh
9475ccc3d6 Fix polynomial computation. 
Fix incorrect block length when doing fixed-block dedupe.
 2012-10-02 00:20:44 +05:30
Moinak Ghosh
0019efbadb Adjust break pattern check mask for closer approximation to average block size. 
Remove unused structure member.
 2012-09-29 23:31:45 +05:30
Moinak Ghosh
24e6f4e629 Switch to multiplicative rolling hash for good distribution properties. 2012-09-29 00:09:49 +05:30
Moinak Ghosh
8f8af7ed6b Update adaptive mode heuristic based on algorithms. 
Remove incorrect check in PPMd decompression code.
More refactoring of variable names.
 2012-09-27 22:29:08 +05:30
Moinak Ghosh
449dc35675 Speed up adaptive modes by using heuristics to select compression algorithm. 
Select similarity percentage based on dedupe block size for effectiveness.
 2012-09-26 19:47:32 +05:30
Moinak Ghosh
333b7b011e Fix check for size reduction in dedupe. 
Tweak debug message for clarity.
 2012-09-25 23:24:33 +05:30
Moinak Ghosh
3544a8c708 Fix polynomial table computation. 
Change hashing and length bias to reduce hashtable bucket collisions.
Add support for user-selectable 60% or 40% similarity for Delta Compression.
Overall slight speedup.
 2012-09-24 22:20:27 +05:30
Moinak Ghosh
8386e72566 Rewrite core dedupe logic to simplify code and improve performance. 
Hashtable based chunk-level deduplication instead of Quicksort.
Fix a corner case bug in Dedupe decompression.
 2012-09-23 14:57:09 +05:30
Moinak Ghosh
99a8e4cd98 Speed up Hash computation for dedupe blocks. 
Add missing initialization of sliding window.
Update help text.
 2012-09-19 20:29:44 +05:30
Moinak Ghosh
e3befd9e16 Add support for Fixed-Block deduplication. 
More refactoring of symbol names.
 2012-09-16 11:12:58 +05:30
Moinak Ghosh
b9355a5dcc Reduce dedupe loop checks for slight speed edge. 
Beginnings of Fixed-block dedupe.
Update variable name for clarity.
 2012-09-15 11:14:58 +05:30
Moinak Ghosh
f3f472b860 Implement K-min-values Sketch for Similarity detection. 2012-09-11 20:26:36 +05:30
Moinak Ghosh
e6f042aaf8 Allow user-specified minimum Dedupe block size. 
Compute similarity sketch only if Delta Compression enabled.
 2012-09-05 22:43:54 +05:30
Moinak Ghosh
560fa85aab Fix secondary sketch computation, some more accuracy in diff detection. 2012-09-04 23:28:02 +05:30
Moinak Ghosh
262566b59a Add xxHash for Rabin block checksums, slightly faster than CRC64. 
Fix missing initialization of character counts table.
Some file reorganization.
 2012-09-02 20:40:32 +05:30
Moinak Ghosh
eda312ce1e Add support for Skein512 and Skein256 checksums 
Import Skein code from NIST CD submission
Make checksum algorithms pluggable
Fix handling of huge buffers (>2GB) in LZP
Cleanup of some buffer sizing code
Speed up CRC64 calculation in dedupe chunking
 2012-08-31 22:36:06 +05:30
Moinak Ghosh
bf149e880d Add LZP Pre-Compression support ported from libbsc. 
Add generic pre-processing wrappers for future support of other pre-processors.
Clean up computation of Rabin block sizes.
Compute Rabin scratch space accurately to avoid RAM wastage.
 2012-08-23 22:58:44 +05:30
Moinak Ghosh
023dcae19a Speed up sort comparator function. 2012-08-17 13:18:50 +05:30
Moinak Ghosh
2dadf411fa Reduce memory consumption and improve performance in dedupe 
Re-introduce crc64 for dedup blocks to avoid wasted memcpy-s
Restructe block array to be an array of pointers allocated on demand
Fix a corner case issue when splitting chunks at a dedup boundary
 2012-08-17 11:03:02 +05:30
Moinak Ghosh
55d0485d34 Improve Rabin computations using an irreducible polynomial 
Slight improvement to similarity computation
A simple mechanism to include DEBUG mode stats
Include stdint for common int types
 2012-08-15 20:13:40 +05:30
Moinak Ghosh
3150bdbed7 Implement secondary sketch based on character counts to refine similarity checksum. 
Proper checksum update for last block.
Update comments.
 2012-08-12 13:06:49 +05:30
Moinak Ghosh
bde917c8e9 Fix handling of compression flags in adaptive mode 
Fix error handling when chunk size is too small for dedupe
Bump version to 0.6
 2012-08-10 10:47:11 +05:30
Moinak Ghosh
f2ffcad2fd Compute and compare Mean sketch cksum to improve similarity comparison 
Fix optflags settings in Makefile
Small optimization in zero RLE encoder to avoid scanning during lookahead
Some minor fixes
 2012-08-09 23:57:24 +05:30
Moinak Ghosh
400d0bfa72 Bias fingerprint value with occurrence counts for a better sketch 
Fix latent bug when calling algo deinit in decompression code
Reduce diff threshold for slightly greater delta encoding
Limit similar buffer size difference for less wasted diffing
Change zlib compression wrapper to use faster deflateReset mechanism
Reduce optimization level for Dedupe code, it goes faster
 2012-08-08 22:40:58 +05:30
Moinak Ghosh
927da81562 Remove unneeded checks in qsort comparator. 2012-08-05 18:58:40 +05:30
Moinak Ghosh
203008def9 Further improve LZMA compression parameters to utilize all the 14 levels. 
Tweak some Rabin parmeters for better reduction with zlib and Bzip2.
 2012-07-30 23:30:13 +05:30
Moinak Ghosh
94563a7ecd Fix buffer size computation when allocating Rabin block array. 
Reduce memory usage of Rabin block array.
Add an SSE optimization for bsdiff.
Move integer hashing function to utils file.
More updates to README.
 2012-07-28 23:55:24 +05:30
Moinak Ghosh
c7cc7b469c Update chunk size computation to reduce memory usage. 
Implement runtime bypass of custom allocator.
Update README.
 2012-07-27 22:03:24 +05:30
Moinak Ghosh
53d4311534 Make LZFX Hash size dynamic. 
Use smaller min rabin block when using fast compression algos.
Add missing check for algo init function return value.
 2012-07-23 21:43:12 +05:30
Moinak Ghosh
8cfd54fe34 Add LZFX Compression support, a very fast lightweight compressor. 
Avoid a branch in the rabin loop.
 2012-07-23 00:15:08 +05:30
Moinak Ghosh
7e14909ad1 Separate initial rabin boundary detection and block splitting for performance. 
Also fix a rare corner case latent bug.
 2012-07-22 21:27:44 +05:30
Moinak Ghosh
962a2cae8a Compress Dedup index only if it is at least 90 bytes to avoid expansion. 
Some minor cleanup.
 2012-07-22 00:00:41 +05:30
Moinak Ghosh
b69dcf4d55 Remove debug statements. 2012-07-20 21:38:39 +05:30
Moinak Ghosh
fd7c7e9a65 Use 4-byte ints for header values instead of 8-byte size_t. 
Use RLE on control data if it reduces the size.
Update some comments.
Use scratch space at end of data chunk, if available.
 2012-07-20 20:53:46 +05:30
Moinak Ghosh
e788eb43b8 Implement Delta Encoding based on modified bsdiff. 
Change to more accurate Sketch value computation approach.
 2012-07-19 21:41:07 +05:30