公开(公告)号：US09152333B1

公开(公告)日：2015-10-06

申请号：US13768191

申请日：2013-02-15

Applicant: NetApp, Inc.

Inventor： Craig Anthony Johnston ,  Vinod Daga ,  Subramaniam Periyagaram ,  Alexander James Gronbach ,  Bharadwaj Vellore Ramesh

IPC: G06F12/00 ,  G06F3/06

CPC classification number: G06F3/0641 ,  G06F3/0608 ,  G06F3/067 ,  G06F3/0673

Abstract: Techniques for a method of estimating deduplication potential are disclosed herein. The method includes steps of selecting randomly a plurality of data blocks from a data set as a sample of the data set, collecting fingerprints of the plurality of data blocks of the sample, identifying duplicates of fingerprints of the sample from the fingerprints of the plurality of data blocks, estimating a total number of unique fingerprints of the data set depending on a total number of the duplicates of fingerprints of the sample based on a probability of fingerprints from the data set colliding in the sample, and determining a total number of duplicates of fingerprints of the data set depending on the total number of the unique fingerprints of the data set.

Abstract translation: 本文中公开了一种估算重复数据消除潜力的方法。 该方法包括以下步骤：从作为数据集的样本的数据集中随机选择多个数据块，收集样本的多个数据块的指纹，从多个数据集的指纹中识别样本的指纹的重复 数据块，基于来自与样本相冲突的数据集的指纹的概率，根据所述样本的指纹的副本的总数来估计所述数据集的唯一指纹的总数，并且确定所述样本的副本的总数 取决于数据集的唯一指纹的总数的数据集的指纹。

公开(公告)号：US20140358873A1

公开(公告)日：2014-12-04

申请号：US14460032

申请日：2014-08-14

Applicant: NETAPP, INC.

Inventor： Vinod Kumar Daga ,  Craig Anthony Johnston ,  Ling Zheng

IPC: G06F17/30

CPC classification number: G06F17/30303 ,  G06F17/30079 ,  G06F17/3015 ,  G06F17/30156

Abstract: A method performed in a system that has a plurality of volumes stored to storage hardware, the method including generating, for each of the volumes, a respective space saving potential iteratively over time and scheduling space saving operations among the plurality of volumes by analyzing each of the volumes for space saving potential and assigning priority of resources based at least in part on space saving potential.

Abstract translation: 一种在具有存储到存储硬件的多个卷的系统中执行的方法，所述方法包括针对每个卷生成随时间迭代的相应的空间节省潜力，并且通过分析每个卷中的每个 节省空间的潜力，并至少部分依赖节省空间的潜力来分配资源的优先级。