- *If total drives has many common divisors the algorithm chooses the minimum amounts of erasure sets possible for a erasure set size of any N*.  In the example with 1024 drives - 4, 8, 16 are GCD factors. With 16 drives we get a total of 64 possible sets, with 8 drives we get a total of 128 possible sets, with 4 drives we get a total of 256 possible sets. So algorithm automatically chooses 64 sets, which is *16* 64 = 1024* drives in total.

		// for unset: 1024. Return 2 here to really match kernel default for
		// zero milliCPU.
		return MinShares
	}
	// Conceptually (milliCPU / milliCPUToCPU) * sharesPerCPU, but factored to improve rounding.
	shares := (milliCPU * SharesPerCPU) / MilliCPUToCPU
	if shares < MinShares {
		return MinShares
	}
	if shares > MaxShares {
		return MaxShares
	}
	return uint64(shares)
}

}

// quotToFloat32 returns the non-negative float32 value
// nearest to the quotient a/b, using round-to-even in
// halfway cases. It does not mutate its arguments.
// Preconditions: b is non-zero; a and b have no common factors.
func quotToFloat32(a, b nat) (f float32, exact bool) {
	const (
		// float size in bits
		Fsize = 32

		// mantissa
		Msize  = 23
		Msize1 = Msize + 1 // incl. implicit 1
		Msize2 = Msize1 + 1

limitations under the License.
*/

package resource

import (
	"math/big"

	inf "gopkg.in/inf.v0"
)

const (
	// maxInt64Factors is the highest value that will be checked when removing factors of 10 from an int64.
	// It is also the maximum decimal digits that can be represented with an int64.
	maxInt64Factors = 18
)

var (
	// Commonly needed big.Int values-- treat as read only!

 * <p>This class should not be assumed to be universally superior to {@code
 * java.util.LinkedHashSet}. Generally speaking, this class reduces object allocation and memory
 * consumption at the price of moderately increased constant factors of CPU. Only use this class
 * when there is a specific reason to prioritize memory over CPU.
 *
 * @author Louis Wasserman
 */
@GwtIncompatible // not worth using in GWT for now
@ElementTypesAreNonnullByDefault

 * <p>This class should not be assumed to be universally superior to {@code
 * java.util.LinkedHashSet}. Generally speaking, this class reduces object allocation and memory
 * consumption at the price of moderately increased constant factors of CPU. Only use this class
 * when there is a specific reason to prioritize memory over CPU.
 *
 * @author Louis Wasserman
 */
@GwtIncompatible // not worth using in GWT for now
@ElementTypesAreNonnullByDefault

	// preemption is given by the formula:
	// numCandidates = max(numNodes * minCandidateNodesPercentage, minCandidateNodesAbsolute)
	// We say "likely" because there are other factors such as PDB violations
	// that play a role in the number of candidates shortlisted. Must be at least
	// 0 nodes. Defaults to 100 nodes if unspecified.
	MinCandidateNodesAbsolute int32
}

 * <p>This class should not be assumed to be universally superior to {@code
 * java.util.LinkedHashMap}. Generally speaking, this class reduces object allocation and memory
 * consumption at the price of moderately increased constant factors of CPU. Only use this class
 * when there is a specific reason to prioritize memory over CPU.
 *
 * @author Louis Wasserman
 */
@J2ktIncompatible // no support for access-order mode in LinkedHashMap delegate

Unit tests are easy to cache as they normally have no external dependencies.
For integration tests the situation can be quite different, as they can depend on a variety of inputs outside of the test and production code.
These external factors can be for example:

  * operating system type and version,
  * external tools being installed for the tests,
  * environment variables and Java system properties,
  * other services being up and running,

option cc_enable_arenas = true;

// This file contains the definition of TF GraphDef-level mixed-precision
// quantization configuration. The configuration will be used in the
// quantization path to determine the following factors:
// 1) What will be the quantization method for the model.
// 2) What will be the default quantization precision for the model.
// 3) What will be the quantization precision for each unit (nodes / ops) in the
//    model.