Files
049323c40f Add Parquet block format (#1479)
* starting parquet support

starting parquet support. some things are still todo:  compaction, find trace by ID

Co-Authored-By: Annanay Agarwal <[email protected]>
Co-Authored-By: Martin Disibio <[email protected]>

* add parquet reusable streaming block

* Add parquet compaction

* Add findTraceByID logic and parquet -> proto conversion

Signed-off-by: Annanay <[email protected]>

* Switch findtracebyid to readrow until we get SeekToRow working

* Fix error handling in bufferedReaderAt. Add various instrumentation and log messages

* Use jsonpb to roundtrip event attributes

* Improve buffered reader when reading near end of file. More instrumentation and tidiness

* Add totalBlockBytes instrumentation. Fix inspectedBytes instrumentation

* parquet.FindTraceByID check bloom

* Make compator flush on trace count until we get the row group size target working

* Support maxDuration in parquet search

* Correct attribute translation for array/kvlist

Signed-off-by: Annanay <[email protected]>

* extend roundtrip test

Signed-off-by: Annanay <[email protected]>

* Use default page buffer size

* Fix size url param to be uint64

* Increase parquet searches to 1MB stripes

* Add test for findtracebyid

* make row group size configurable

Signed-off-by: Annanay <[email protected]>

* Record parquet footer size in block meta and short-circuit header/footer reads

* Add tests for SubstringPredicate

* Switch compaction to compact blocks of same version, use re-usable buffer in findtracebyid

Signed-off-by: Annanay <[email protected]>

* Add search support for remaining well-known attributes and span status, add search tests

* Benchmark substringPredicate

* Make block version configurable, default to v2

* memoize substringPredicate

* Update segmentio/parquet-go to catch SeekToRow fix, revert hack in FindTraceByID

Signed-off-by: Annanay <[email protected]>

* Add multi block prefetch iterator for use in compactor

Signed-off-by: Annanay <[email protected]>

* Add BufferedWriter with outgoing queue to flush traces async

Signed-off-by: Annanay <[email protected]>

Make compactor block complete async

Signed-off-by: Annanay <[email protected]>

* First pass CLI to dump schema and column sizes from block

Signed-off-by: Annanay <[email protected]>

* honor search start/end time range

* Improve error message, add blockID

Signed-off-by: Annanay <[email protected]>

* Fix race in InstrumentedPredicate with atomics

* Make bufferedReaderAt not single-threaded

* Debug SeekToRow panic, add test data from vulture tenant

Signed-off-by: Annanay <[email protected]>

* fix test

* repair tests

* Address println warnings

* Experiment with bookmark prefetch

Signed-off-by: Annanay <[email protected]>

Remove quitCh from bookmark

Signed-off-by: Annanay <[email protected]>

* fix prefetch on the bookmarks

Signed-off-by: Annanay <[email protected]>

* lint

* tracker -> RowNumber, lint

* delete unused predicates

* lint

* lint

* mod vendor

* remove pool from buffered writer queue

Signed-off-by: Annanay <[email protected]>

* post merge cleanup

Signed-off-by: Annanay <[email protected]>

* Close quitCh instead of attempting to signal to it

Signed-off-by: Annanay <[email protected]>

* split multiblock and prefetch iterators, refactor around interface

Signed-off-by: Annanay <[email protected]>

* Split search into 2 phases to avoid loading traceid/display columns unless there is a match

* Make columnIterator.SeekTo do something

* update to catch segmentio/parquet-go#220

Signed-off-by: Annanay <[email protected]>

* Fix column index to store entire 32-byte trace ID string. lint

* Make parquet honor the adjusted start/end times that come from the wal

* post merge cleanup: use compaction metric callbacks

Signed-off-by: Annanay <[email protected]>

* serverless sum/mods

Signed-off-by: Annanay <[email protected]>

* parquetquery predicates handle column index not present

* Make parquet compaction really honor input block start/end times, add universal test at tempodb-level

* parquet skip page indexes while searching to reduce latency on openfile

* More testing around iterators and combiner

Signed-off-by: Annanay <[email protected]>

* Update to latest parquet lib to catch rle bitpack fix

* Review feedback:  LatestEncoding -> DefaultEncoding

* Pass block footerSize through serverless

* mod vendor for serverless

* Fix tests

* add blockID to error message in block iterator

Signed-off-by: Annanay <[email protected]>

* increase readvalues buffer, add more span instrumentation

Signed-off-by: Annanay <[email protected]>

* changelog and docs

* Remove the use of async writer in vparquet compactor

Signed-off-by: Annanay <[email protected]>

Co-authored-by: Annanay Agarwal <[email protected]>
2022-06-23 16:00:18 +05:30

183 lines
4.6 KiB
Go

package brotli
func utf8Position(last uint, c uint, clamp uint) uint {
if c < 128 {
return 0 /* Next one is the 'Byte 1' again. */
} else if c >= 192 { /* Next one is the 'Byte 2' of utf-8 encoding. */
return brotli_min_size_t(1, clamp)
} else {
/* Let's decide over the last byte if this ends the sequence. */
if last < 0xE0 {
return 0 /* Completed two or three byte coding. */ /* Next one is the 'Byte 3' of utf-8 encoding. */
} else {
return brotli_min_size_t(2, clamp)
}
}
}
func decideMultiByteStatsLevel(pos uint, len uint, mask uint, data []byte) uint {
var counts = [3]uint{0} /* should be 2, but 1 compresses better. */
var max_utf8 uint = 1
var last_c uint = 0
var i uint
for i = 0; i < len; i++ {
var c uint = uint(data[(pos+i)&mask])
counts[utf8Position(last_c, c, 2)]++
last_c = c
}
if counts[2] < 500 {
max_utf8 = 1
}
if counts[1]+counts[2] < 25 {
max_utf8 = 0
}
return max_utf8
}
func estimateBitCostsForLiteralsUTF8(pos uint, len uint, mask uint, data []byte, cost []float32) {
var max_utf8 uint = decideMultiByteStatsLevel(pos, uint(len), mask, data)
/* Bootstrap histograms. */
var histogram = [3][256]uint{[256]uint{0}}
var window_half uint = 495
var in_window uint = brotli_min_size_t(window_half, uint(len))
var in_window_utf8 = [3]uint{0}
/* max_utf8 is 0 (normal ASCII single byte modeling),
1 (for 2-byte UTF-8 modeling), or 2 (for 3-byte UTF-8 modeling). */
var i uint
{
var last_c uint = 0
var utf8_pos uint = 0
for i = 0; i < in_window; i++ {
var c uint = uint(data[(pos+i)&mask])
histogram[utf8_pos][c]++
in_window_utf8[utf8_pos]++
utf8_pos = utf8Position(last_c, c, max_utf8)
last_c = c
}
}
/* Compute bit costs with sliding window. */
for i = 0; i < len; i++ {
if i >= window_half {
var c uint
var last_c uint
if i < window_half+1 {
c = 0
} else {
c = uint(data[(pos+i-window_half-1)&mask])
}
if i < window_half+2 {
last_c = 0
} else {
last_c = uint(data[(pos+i-window_half-2)&mask])
}
/* Remove a byte in the past. */
var utf8_pos2 uint = utf8Position(last_c, c, max_utf8)
histogram[utf8_pos2][data[(pos+i-window_half)&mask]]--
in_window_utf8[utf8_pos2]--
}
if i+window_half < len {
var c uint = uint(data[(pos+i+window_half-1)&mask])
var last_c uint = uint(data[(pos+i+window_half-2)&mask])
/* Add a byte in the future. */
var utf8_pos2 uint = utf8Position(last_c, c, max_utf8)
histogram[utf8_pos2][data[(pos+i+window_half)&mask]]++
in_window_utf8[utf8_pos2]++
}
{
var c uint
var last_c uint
if i < 1 {
c = 0
} else {
c = uint(data[(pos+i-1)&mask])
}
if i < 2 {
last_c = 0
} else {
last_c = uint(data[(pos+i-2)&mask])
}
var utf8_pos uint = utf8Position(last_c, c, max_utf8)
var masked_pos uint = (pos + i) & mask
var histo uint = histogram[utf8_pos][data[masked_pos]]
var lit_cost float64
if histo == 0 {
histo = 1
}
lit_cost = fastLog2(in_window_utf8[utf8_pos]) - fastLog2(histo)
lit_cost += 0.02905
if lit_cost < 1.0 {
lit_cost *= 0.5
lit_cost += 0.5
}
/* Make the first bytes more expensive -- seems to help, not sure why.
Perhaps because the entropy source is changing its properties
rapidly in the beginning of the file, perhaps because the beginning
of the data is a statistical "anomaly". */
if i < 2000 {
lit_cost += 0.7 - (float64(2000-i) / 2000.0 * 0.35)
}
cost[i] = float32(lit_cost)
}
}
}
func estimateBitCostsForLiterals(pos uint, len uint, mask uint, data []byte, cost []float32) {
if isMostlyUTF8(data, pos, mask, uint(len), kMinUTF8Ratio) {
estimateBitCostsForLiteralsUTF8(pos, uint(len), mask, data, cost)
return
} else {
var histogram = [256]uint{0}
var window_half uint = 2000
var in_window uint = brotli_min_size_t(window_half, uint(len))
var i uint
/* Bootstrap histogram. */
for i = 0; i < in_window; i++ {
histogram[data[(pos+i)&mask]]++
}
/* Compute bit costs with sliding window. */
for i = 0; i < len; i++ {
var histo uint
if i >= window_half {
/* Remove a byte in the past. */
histogram[data[(pos+i-window_half)&mask]]--
in_window--
}
if i+window_half < len {
/* Add a byte in the future. */
histogram[data[(pos+i+window_half)&mask]]++
in_window++
}
histo = histogram[data[(pos+i)&mask]]
if histo == 0 {
histo = 1
}
{
var lit_cost float64 = fastLog2(in_window) - fastLog2(histo)
lit_cost += 0.029
if lit_cost < 1.0 {
lit_cost *= 0.5
lit_cost += 0.5
}
cost[i] = float32(lit_cost)
}
}
}
}