private static void startCDC() throws IOException, Exception {
long remainder = 7;
for (int i = startBoundary; i <= endBoundary; i += increment) {
long minBoundary = min_multiplier * i; // we will set the mod value as the minimum boundary
long maxBoundary = max_multiplier * i; // we will set this as the maximum boundary
long divisor1 = i; // this will be used to mod the results
long divisor2 = i / 2 + 1; // the backup divisor is half the original divisor
long divisor3 = i / 4 + 1;
totalSize =
fileArray.get(1)
.length; // note we only care about the size of the second file since that's the file
// we are measuring
System.out.print(divisor1 + " " + divisor2 + " " + divisor3 + " ");
runBytes(
window,
divisor1,
divisor2,
divisor3,
remainder,
minBoundary,
maxBoundary); // run the karb rabin algorithm
// this is the block size per boundary
double blockSize = (double) totalSize / (double) numOfPieces;
double ratio = (double) coverage / (double) totalSize;
System.out.println(blockSize + " " + ratio);
// clear the hashTable, and counters so we can reset the values for the next round of
// boundaries
coverage = 0;
numOfPieces = 0;
table.clear();
HashClass.duplicate_counter = 0;
HashClass.max_list_length = 0;
}
}
/*
- Overloaded method just for the internet archive dataset
- The first two params hold the block size and ratioSize respectively (for all the runnings)
- The last set of params are the actual file in byte and the hashed versions of the file we will be running the code against
- current_ -- are the lists that contain the most recent file version
- previous_ -- are the listrs that contain the previous versions
*/
private static void startCDC(
double[] block_size_list,
double[] ratio_size_list,
byte[] current_array,
byte[] previous_array,
ArrayList<Long> current_md5Hashes,
ArrayList<Long> previous_md5Hashes)
throws Exception {
long remainder = 7; // this is the remainder that we will be comparing with
int index = 0; // used to traverse the two lists
for (int i = startBoundary; i <= endBoundary; i += increment) {
long minBoundary = min_multiplier * i; // we will set the mod value as the minimum boundary
long maxBoundary = max_multiplier * i; // we will set this as the maximum boundary
long divisor1 = i; // this will be used to mod the results
long divisor2 = i / 2 + 1; // the backup divisor is half the original divisor
long divisor3 = i / 4 + 1;
// System.out.print( i+" ");
storeChunks(
previous_array,
previous_md5Hashes,
divisor1,
divisor2,
divisor3,
remainder,
minBoundary,
maxBoundary); // cut up the first file and store it
runTddd(
current_array,
current_md5Hashes,
divisor1,
divisor2,
divisor3,
remainder,
minBoundary,
maxBoundary); // call the method again, but on the second file only
// this is the block size per boundary
double blockSize = (double) totalSize / (double) numOfPieces;
double ratio = (double) coverage / (double) totalSize;
// extra step, add the data back into the list
block_size_list[index] += blockSize;
ratio_size_list[index] += ratio;
++index;
// clear the hashTable, and counters so we can reset the values for the next round of
// boundaries
coverage = 0;
numOfPieces = 0;
table.clear();
HashClass.duplicate_counter = 0;
HashClass.max_list_length = 0;
}
}