Parity API (#280)

This commit is contained in:
Daniel Valdivia
2020-09-22 11:15:21 -07:00
committed by GitHub
parent c0ee739624
commit e5f7870f5e
13 changed files with 1321 additions and 0 deletions

219
pkg/utils/parity.go Normal file
View File

@@ -0,0 +1,219 @@
// This file is part of MinIO Console Server
// Copyright (c) 2020 MinIO, Inc.
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU Affero General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Affero General Public License for more details.
//
// You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>.
package utils
import (
"errors"
"fmt"
"sort"
"github.com/minio/minio/pkg/ellipses"
)
// This file implements and supports ellipses pattern for
// `minio server` command line arguments.
// Supported set sizes this is used to find the optimal
// single set size.
var setSizes = []uint64{4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16}
// getDivisibleSize - returns a greatest common divisor of
// all the ellipses sizes.
func getDivisibleSize(totalSizes []uint64) (result uint64) {
gcd := func(x, y uint64) uint64 {
for y != 0 {
x, y = y, x%y
}
return x
}
result = totalSizes[0]
for i := 1; i < len(totalSizes); i++ {
result = gcd(result, totalSizes[i])
}
return result
}
// isValidSetSize - checks whether given count is a valid set size for erasure coding.
var isValidSetSize = func(count uint64) bool {
return (count >= setSizes[0] && count <= setSizes[len(setSizes)-1])
}
// possibleSetCountsWithSymmetry returns symmetrical setCounts based on the
// input argument patterns, the symmetry calculation is to ensure that
// we also use uniform number of drives common across all ellipses patterns.
func possibleSetCountsWithSymmetry(setCounts []uint64, argPatterns []ellipses.ArgPattern) []uint64 {
var newSetCounts = make(map[uint64]struct{})
for _, ss := range setCounts {
var symmetry bool
for _, argPattern := range argPatterns {
for _, p := range argPattern {
if uint64(len(p.Seq)) > ss {
symmetry = uint64(len(p.Seq))%ss == 0
} else {
symmetry = ss%uint64(len(p.Seq)) == 0
}
}
}
// With no arg patterns, it is expected that user knows
// the right symmetry, so either ellipses patterns are
// provided (recommended) or no ellipses patterns.
if _, ok := newSetCounts[ss]; !ok && (symmetry || argPatterns == nil) {
newSetCounts[ss] = struct{}{}
}
}
setCounts = []uint64{}
for setCount := range newSetCounts {
setCounts = append(setCounts, setCount)
}
// Not necessarily needed but it ensures to the readers
// eyes that we prefer a sorted setCount slice for the
// subsequent function to figure out the right common
// divisor, it avoids loops.
sort.Slice(setCounts, func(i, j int) bool {
return setCounts[i] < setCounts[j]
})
return setCounts
}
func commonSetDriveCount(divisibleSize uint64, setCounts []uint64) (setSize uint64) {
// prefers setCounts to be sorted for optimal behavior.
if divisibleSize < setCounts[len(setCounts)-1] {
return divisibleSize
}
// Figure out largest value of total_drives_in_erasure_set which results
// in least number of total_drives/total_drives_erasure_set ratio.
prevD := divisibleSize / setCounts[0]
for _, cnt := range setCounts {
if divisibleSize%cnt == 0 {
d := divisibleSize / cnt
if d <= prevD {
prevD = d
setSize = cnt
}
}
}
return setSize
}
// getSetIndexes returns list of indexes which provides the set size
// on each index, this function also determines the final set size
// The final set size has the affinity towards choosing smaller
// indexes (total sets)
func getSetIndexes(args []string, totalSizes []uint64, argPatterns []ellipses.ArgPattern) (setIndexes [][]uint64, err error) {
if len(totalSizes) == 0 || len(args) == 0 {
return nil, errors.New("invalid argument")
}
setIndexes = make([][]uint64, len(totalSizes))
for _, totalSize := range totalSizes {
// Check if totalSize has minimum range upto setSize
if totalSize < setSizes[0] {
return nil, fmt.Errorf("incorrect number of endpoints provided %s", args)
}
}
commonSize := getDivisibleSize(totalSizes)
possibleSetCounts := func(setSize uint64) (ss []uint64) {
for _, s := range setSizes {
if setSize%s == 0 {
ss = append(ss, s)
}
}
return ss
}
setCounts := possibleSetCounts(commonSize)
if len(setCounts) == 0 {
err = fmt.Errorf("incorrect number of endpoints provided %s, number of disks %d is not divisible by any supported erasure set sizes %d", args, commonSize, setSizes)
return nil, err
}
// Returns possible set counts with symmetry.
setCounts = possibleSetCountsWithSymmetry(setCounts, argPatterns)
if len(setCounts) == 0 {
err = fmt.Errorf("no symmetric distribution detected with input endpoints provided %s, disks %d cannot be spread symmetrically by any supported erasure set sizes %d", args, commonSize, setSizes)
return nil, err
}
// Final set size with all the symmetry accounted for.
setSize := commonSetDriveCount(commonSize, setCounts)
// Check whether setSize is with the supported range.
if !isValidSetSize(setSize) {
err = fmt.Errorf("incorrect number of endpoints provided %s, number of disks %d is not divisible by any supported erasure set sizes %d", args, commonSize, setSizes)
return nil, err
}
for i := range totalSizes {
for j := uint64(0); j < totalSizes[i]/setSize; j++ {
setIndexes[i] = append(setIndexes[i], setSize)
}
}
return setIndexes, nil
}
// Return the total size for each argument patterns.
func getTotalSizes(argPatterns []ellipses.ArgPattern) []uint64 {
var totalSizes []uint64
for _, argPattern := range argPatterns {
var totalSize uint64 = 1
for _, p := range argPattern {
totalSize = totalSize * uint64(len(p.Seq))
}
totalSizes = append(totalSizes, totalSize)
}
return totalSizes
}
// PossibleParityValues returns possible parities for input args,
// parties are calculated in uniform manner for one zone or
// multiple zones, ensuring that parities returned are common
// and applicable across all zones.
func PossibleParityValues(args ...string) ([]string, error) {
setIndexes, err := parseEndpointSet(args...)
if err != nil {
return nil, err
}
maximumParity := setIndexes[0][0] / 2
var parities []string
for maximumParity >= 2 {
parities = append(parities, fmt.Sprintf("EC:%d", maximumParity))
maximumParity--
}
return parities, nil
}
// Parses all arguments and returns an endpointSet which is a collection
// of endpoints following the ellipses pattern, this is what is used
// by the object layer for initializing itself.
func parseEndpointSet(args ...string) (setIndexes [][]uint64, err error) {
var argPatterns = make([]ellipses.ArgPattern, len(args))
for i, arg := range args {
patterns, err := ellipses.FindEllipsesPatterns(arg)
if err != nil {
return nil, err
}
argPatterns[i] = patterns
}
return getSetIndexes(args, getTotalSizes(argPatterns), argPatterns)
}

281
pkg/utils/parity_test.go Normal file
View File

@@ -0,0 +1,281 @@
// This file is part of MinIO Console Server
// Copyright (c) 2020 MinIO, Inc.
//
// All rights reserved
package utils
import (
"reflect"
"testing"
"github.com/minio/minio/pkg/ellipses"
)
func TestGetDivisibleSize(t *testing.T) {
testCases := []struct {
totalSizes []uint64
result uint64
}{{[]uint64{24, 32, 16}, 8},
{[]uint64{32, 8, 4}, 4},
{[]uint64{8, 8, 8}, 8},
{[]uint64{24}, 24},
}
for _, testCase := range testCases {
testCase := testCase
t.Run("", func(t *testing.T) {
gotGCD := getDivisibleSize(testCase.totalSizes)
if testCase.result != gotGCD {
t.Errorf("Expected %v, got %v", testCase.result, gotGCD)
}
})
}
}
// Test tests calculating set indexes.
func TestGetSetIndexes(t *testing.T) {
testCases := []struct {
args []string
totalSizes []uint64
indexes [][]uint64
success bool
}{
// Invalid inputs.
{
[]string{"data{1...3}"},
[]uint64{3},
nil,
false,
},
{
[]string{"data/controller1/export{1...2}, data/controller2/export{1...4}, data/controller3/export{1...8}"},
[]uint64{2, 4, 8},
nil,
false,
},
{
[]string{"data{1...17}/export{1...52}"},
[]uint64{14144},
nil,
false,
},
// Valid inputs.
{
[]string{"data{1...27}"},
[]uint64{27},
[][]uint64{{9, 9, 9}},
true,
},
{
[]string{"http://host{1...3}/data{1...180}"},
[]uint64{540},
[][]uint64{{15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15}},
true,
},
{
[]string{"http://host{1...2}.rack{1...4}/data{1...180}"},
[]uint64{1440},
[][]uint64{{16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16}},
true,
},
{
[]string{"http://host{1...2}/data{1...180}"},
[]uint64{360},
[][]uint64{{12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12}},
true,
},
{
[]string{"data/controller1/export{1...4}, data/controller2/export{1...8}, data/controller3/export{1...12}"},
[]uint64{4, 8, 12},
[][]uint64{{4}, {4, 4}, {4, 4, 4}},
true,
},
{
[]string{"data{1...64}"},
[]uint64{64},
[][]uint64{{16, 16, 16, 16}},
true,
},
{
[]string{"data{1...24}"},
[]uint64{24},
[][]uint64{{12, 12}},
true,
},
{
[]string{"data/controller{1...11}/export{1...8}"},
[]uint64{88},
[][]uint64{{11, 11, 11, 11, 11, 11, 11, 11}},
true,
},
{
[]string{"data{1...4}"},
[]uint64{4},
[][]uint64{{4}},
true,
},
{
[]string{"data/controller1/export{1...10}, data/controller2/export{1...10}, data/controller3/export{1...10}"},
[]uint64{10, 10, 10},
[][]uint64{{10}, {10}, {10}},
true,
},
{
[]string{"data{1...16}/export{1...52}"},
[]uint64{832},
[][]uint64{{16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16}},
true,
},
}
for _, testCase := range testCases {
testCase := testCase
t.Run("", func(t *testing.T) {
var argPatterns = make([]ellipses.ArgPattern, len(testCase.args))
for i, arg := range testCase.args {
patterns, err := ellipses.FindEllipsesPatterns(arg)
if err != nil {
t.Fatalf("Unexpected failure %s", err)
}
argPatterns[i] = patterns
}
gotIndexes, err := getSetIndexes(testCase.args, testCase.totalSizes, argPatterns)
if err != nil && testCase.success {
t.Errorf("Expected success but failed instead %s", err)
}
if err == nil && !testCase.success {
t.Errorf("Expected failure but passed instead")
}
if !reflect.DeepEqual(testCase.indexes, gotIndexes) {
t.Errorf("Expected %v, got %v", testCase.indexes, gotIndexes)
}
})
}
}
// Test tests possible parities returned for any input args
func TestPossibleParities(t *testing.T) {
testCases := []struct {
arg string
parities []string
success bool
}{
// Tests invalid inputs.
{
"...",
nil,
false,
},
// No range specified.
{
"{...}",
nil,
false,
},
// Invalid range.
{
"http://minio{2...3}/export/set{1...0}",
nil,
false,
},
// Range cannot be smaller than 4 minimum.
{
"/export{1..2}",
nil,
false,
},
// Unsupported characters.
{
"/export/test{1...2O}",
nil,
false,
},
// Tests valid inputs.
{
"{1...27}",
[]string{"EC:4", "EC:3", "EC:2"},
true,
},
{
"/export/set{1...64}",
[]string{"EC:8", "EC:7", "EC:6", "EC:5", "EC:4", "EC:3", "EC:2"},
true,
},
// Valid input for distributed setup.
{
"http://minio{2...3}/export/set{1...64}",
[]string{"EC:8", "EC:7", "EC:6", "EC:5", "EC:4", "EC:3", "EC:2"},
true,
},
// Supporting some advanced cases.
{
"http://minio{1...64}.mydomain.net/data",
[]string{"EC:8", "EC:7", "EC:6", "EC:5", "EC:4", "EC:3", "EC:2"},
true,
},
{
"http://rack{1...4}.mydomain.minio{1...16}/data",
[]string{"EC:8", "EC:7", "EC:6", "EC:5", "EC:4", "EC:3", "EC:2"},
true,
},
// Supporting kubernetes cases.
{
"http://minio{0...15}.mydomain.net/data{0...1}",
[]string{"EC:8", "EC:7", "EC:6", "EC:5", "EC:4", "EC:3", "EC:2"},
true,
},
// No host regex, just disks.
{
"http://server1/data{1...32}",
[]string{"EC:8", "EC:7", "EC:6", "EC:5", "EC:4", "EC:3", "EC:2"},
true,
},
// No host regex, just disks with two position numerics.
{
"http://server1/data{01...32}",
[]string{"EC:8", "EC:7", "EC:6", "EC:5", "EC:4", "EC:3", "EC:2"},
true,
},
// More than 2 ellipses are supported as well.
{
"http://minio{2...3}/export/set{1...64}/test{1...2}",
[]string{"EC:8", "EC:7", "EC:6", "EC:5", "EC:4", "EC:3", "EC:2"},
true,
},
// More than 1 ellipses per argument for standalone setup.
{
"/export{1...10}/disk{1...10}",
[]string{"EC:5", "EC:4", "EC:3", "EC:2"},
true,
},
// IPv6 ellipses with hexadecimal expansion
{
"http://[2001:3984:3989::{1...a}]/disk{1...10}",
[]string{"EC:5", "EC:4", "EC:3", "EC:2"},
true,
},
// IPv6 ellipses with hexadecimal expansion with 3 position numerics.
{
"http://[2001:3984:3989::{001...00a}]/disk{1...10}",
[]string{"EC:5", "EC:4", "EC:3", "EC:2"},
true,
},
}
for _, testCase := range testCases {
testCase := testCase
t.Run("", func(t *testing.T) {
gotPs, err := PossibleParityValues(testCase.arg)
if err != nil && testCase.success {
t.Errorf("Expected success but failed instead %s", err)
}
if err == nil && !testCase.success {
t.Errorf("Expected failure but passed instead")
}
if !reflect.DeepEqual(testCase.parities, gotPs) {
t.Errorf("Expected %v, got %v", testCase.parities, gotPs)
}
})
}
}