Add support for Uncompress(source, sink). Various changes to allow

Uncompress(source, sink) to get the same performance as the different
variants of Uncompress to Cord/DataBuffer/String/FlatBuffer.

Changes to efficiently support Uncompress(source, sink)
--------

a) For strings - we add support to StringByteSink to do GetAppendBuffer so we
   can write to it without copying.
b) For flat array buffers, we do GetAppendBuffer and see if we can get a full buffer.

With the above changes we get performance with ByteSource/ByteSink
that is	very close to directly using flat arrays and strings.

We add various benchmark cases to demonstrate that.

Orthogonal change
------------------

Add support for TryFastAppend() for SnappyScatteredWriter.

Benchmark results are below

CPU: Intel Core2 dL1:32KB dL2:4096KB
Benchmark              Time(ns)    CPU(ns) Iterations
-----------------------------------------------------
BM_UFlat/0               109065     108996       6410 896.0MB/s  html
BM_UFlat/1              1012175    1012343        691 661.4MB/s  urls
BM_UFlat/2                26775      26771      26149 4.4GB/s  jpg
BM_UFlat/3                48947      48940      14363 1.8GB/s  pdf
BM_UFlat/4               441029     440835       1589 886.1MB/s  html4
BM_UFlat/5                39861      39880      17823 588.3MB/s  cp
BM_UFlat/6                18315      18300      38126 581.1MB/s  c
BM_UFlat/7                 5254       5254     100000 675.4MB/s  lsp
BM_UFlat/8              1568060    1567376        447 626.6MB/s  xls
BM_UFlat/9               337512     337734       2073 429.5MB/s  txt1
BM_UFlat/10              287269     287054       2434 415.9MB/s  txt2
BM_UFlat/11              890098     890219        787 457.2MB/s  txt3
BM_UFlat/12             1186593    1186863        590 387.2MB/s  txt4
BM_UFlat/13              573927     573318       1000 853.7MB/s  bin
BM_UFlat/14               64250      64294      10000 567.2MB/s  sum
BM_UFlat/15                7301       7300      96153 552.2MB/s  man
BM_UFlat/16              109617     109636       6375 1031.5MB/s  pb
BM_UFlat/17              364438     364497       1921 482.3MB/s  gaviota
BM_UFlatSink/0           108518     108465       6450 900.4MB/s  html
BM_UFlatSink/1           991952     991997        705 675.0MB/s  urls
BM_UFlatSink/2            26815      26798      26065 4.4GB/s  jpg
BM_UFlatSink/3            49127      49122      14255 1.8GB/s  pdf
BM_UFlatSink/4           436674     436731       1604 894.4MB/s  html4
BM_UFlatSink/5            39738      39733      17345 590.5MB/s  cp
BM_UFlatSink/6            18413      18416      37962 577.4MB/s  c
BM_UFlatSink/7             5677       5676     100000 625.2MB/s  lsp
BM_UFlatSink/8          1552175    1551026        451 633.2MB/s  xls
BM_UFlatSink/9           338526     338489       2065 428.5MB/s  txt1
BM_UFlatSink/10          289387     289307       2420 412.6MB/s  txt2
BM_UFlatSink/11          893803     893706        783 455.4MB/s  txt3
BM_UFlatSink/12         1195919    1195459        586 384.4MB/s  txt4
BM_UFlatSink/13          559637     559779       1000 874.3MB/s  bin
BM_UFlatSink/14           65073      65094      10000 560.2MB/s  sum
BM_UFlatSink/15            7618       7614      92823 529.5MB/s  man
BM_UFlatSink/16          110085     110121       6352 1027.0MB/s  pb
BM_UFlatSink/17          369196     368915       1896 476.5MB/s  gaviota
BM_UValidate/0            46954      46957      14899 2.0GB/s  html
BM_UValidate/1           500621     500868       1000 1.3GB/s  urls
BM_UValidate/2              283        283    2481447 417.2GB/s  jpg
BM_UValidate/3            16230      16228      43137 5.4GB/s  pdf
BM_UValidate/4           189129     189193       3701 2.0GB/s  html4

A=uday
R=sanjay
This commit is contained in:
Steinar H. Gunderson
2015-06-22 16:03:28 +02:00
parent b2ad960067
commit b2312c4c25
5 changed files with 420 additions and 1 deletions

View File

@@ -488,6 +488,23 @@ static int VerifyString(const string& input) {
return uncompressed.size();
}
static void VerifyStringSink(const string& input) {
string compressed;
DataEndingAtUnreadablePage i(input);
const size_t written = snappy::Compress(i.data(), i.size(), &compressed);
CHECK_EQ(written, compressed.size());
CHECK_LE(compressed.size(),
snappy::MaxCompressedLength(input.size()));
CHECK(snappy::IsValidCompressedBuffer(compressed.data(), compressed.size()));
string uncompressed;
uncompressed.resize(input.size());
snappy::UncheckedByteArraySink sink(string_as_array(&uncompressed));
DataEndingAtUnreadablePage c(compressed);
snappy::ByteArraySource source(c.data(), c.size());
CHECK(snappy::Uncompress(&source, &sink));
CHECK_EQ(uncompressed, input);
}
static void VerifyIOVec(const string& input) {
string compressed;
@@ -559,6 +576,28 @@ static void VerifyNonBlockedCompression(const string& input) {
CHECK(snappy::Uncompress(compressed.data(), compressed.size(), &uncomp_str));
CHECK_EQ(uncomp_str, input);
// Uncompress using source/sink
string uncomp_str2;
uncomp_str2.resize(input.size());
snappy::UncheckedByteArraySink sink(string_as_array(&uncomp_str2));
snappy::ByteArraySource source(compressed.data(), compressed.size());
CHECK(snappy::Uncompress(&source, &sink));
CHECK_EQ(uncomp_str2, input);
// Uncompress into iovec
{
static const int kNumBlocks = 10;
struct iovec vec[kNumBlocks];
const int block_size = 1 + input.size() / kNumBlocks;
string iovec_data(block_size * kNumBlocks, 'x');
for (int i = 0; i < kNumBlocks; i++) {
vec[i].iov_base = string_as_array(&iovec_data) + i * block_size;
vec[i].iov_len = block_size;
}
CHECK(snappy::RawUncompressToIOVec(compressed.data(), compressed.size(),
vec, kNumBlocks));
CHECK_EQ(string(iovec_data.data(), input.size()), input);
}
}
// Expand the input so that it is at least K times as big as block size
@@ -577,6 +616,8 @@ static int Verify(const string& input) {
// Compress using string based routines
const int result = VerifyString(input);
// Verify using sink based routines
VerifyStringSink(input);
VerifyNonBlockedCompression(input);
VerifyIOVec(input);
@@ -1291,6 +1332,37 @@ static void BM_UIOVec(int iters, int arg) {
}
BENCHMARK(BM_UIOVec)->DenseRange(0, 4);
static void BM_UFlatSink(int iters, int arg) {
StopBenchmarkTiming();
// Pick file to process based on "arg"
CHECK_GE(arg, 0);
CHECK_LT(arg, ARRAYSIZE(files));
string contents = ReadTestDataFile(files[arg].filename,
files[arg].size_limit);
string zcontents;
snappy::Compress(contents.data(), contents.size(), &zcontents);
char* dst = new char[contents.size()];
SetBenchmarkBytesProcessed(static_cast<int64>(iters) *
static_cast<int64>(contents.size()));
SetBenchmarkLabel(files[arg].label);
StartBenchmarkTiming();
while (iters-- > 0) {
snappy::ByteArraySource source(zcontents.data(), zcontents.size());
snappy::UncheckedByteArraySink sink(dst);
CHECK(snappy::Uncompress(&source, &sink));
}
StopBenchmarkTiming();
string s(dst, contents.size());
CHECK_EQ(contents, s);
delete[] dst;
}
BENCHMARK(BM_UFlatSink)->DenseRange(0, ARRAYSIZE(files) - 1);
static void BM_ZFlat(int iters, int arg) {
StopBenchmarkTiming();