公开(公告)号：US07991905B1

公开(公告)日：2011-08-02

申请号：US10365074

申请日：2003-02-12

Applicant: Konstantinos Roussos ,  Yasuhiro Endo ,  Sandeep Yadav

Inventor： Konstantinos Roussos ,  Yasuhiro Endo ,  Sandeep Yadav

IPC: G06F15/16

CPC classification number: H04L65/608 ,  H04L29/08729 ,  H04L65/80 ,  H04L69/28

Abstract: The present invention provides a streaming media cache configured to select a timing interval, chosen from a set of one or more predetermined timing intervals, for each client that requests a data stream from the streaming media cache. Illustratively, clients requesting data streams below a predetermined bit rate (“bandwidth”) are assigned longer timing intervals than clients requesting data streams above the predetermined bit rate. More specifically, because the server can accommodate a larger number of relatively low-bandwidth clients than relatively high-bandwidth clients, the server selects timing intervals of longer duration for the relatively low-bandwidth clients so it may have more time to schedule send times for their requested packets.

Abstract translation: 本发明提供一种流媒体缓存，其被配置为从为流媒体高速缓存请求数据流的每个客户端选择从一组或多个预定定时间隔中选择的定时间隔。 说明性地，请求数据流低于预定比特率（“带宽”）的客户端被分配比客户端更长的定时间隔，该客户端请求高于预定比特率的数据流。 更具体地说，由于服务器可以容纳比相对较高带宽的客户端更多数量的相对低带宽的客户端，所以服务器为相对低带宽的客户端选择较长持续时间的定时间隔，因此可以有更多的时间调度发送时间 他们要求的数据包。

公开(公告)号：US07949865B1

公开(公告)日：2011-05-24

申请号：US11857650

申请日：2007-09-19

Applicant: Sandeep Yadav

Inventor： Sandeep Yadav

IPC: G06F9/24

CPC classification number: G06F17/30197

Abstract: A method for mounting volumes to a storage server in a storage system, including, speeding up initialization of the storage server by staging volume mounting in two or more stages. The method further includes mounting at least one of the volumes when a request for accessing the volume is received.

Abstract translation: 一种用于将卷安装到存储系统中的存储服务器的方法，包括通过在两个或多个阶段分段卷安装来加速存储服务器的初始化。 该方法还包括当接收到访问卷的请求时安装至少一个卷。

公开(公告)号：US20100274773A1

公开(公告)日：2010-10-28

申请号：US12492069

申请日：2009-06-25

Applicant: Dnyaneshwar Pawar ,  Subramaniam V. Periyagaram ,  Sandeep Yadav

Inventor： Dnyaneshwar Pawar ,  Subramaniam V. Periyagaram ,  Sandeep Yadav

IPC: G06F7/00 ,  G06F12/00

CPC classification number: H04L69/04 ,  G06F12/0804 ,  G06F12/0868 ,  G06F17/27 ,  G06F17/30867 ,  G06F2212/2022 ,  G06F2212/214 ,  G06F2212/314 ,  G06F2212/401 ,  H03M7/30

Abstract: A storage server is configured to receive a request to store a data block from a client. The request to store the data block is serviced by the storage server by compressing the data block into a compression group which includes a number of compressed data blocks. The storage server stores the compression group in a non-volatile memory and flushes the compression group from the non-volatile memory to a physical storage device in response to reaching a consistency point. By compressing data to be stored in system memory of a storage servers the amount of data that can be processed during a given time period by a data storage system is increased. Furthermore, an increase in performance can be achieved at a lower cost, since the cost of additional physical system memory modules can be avoided,

Abstract translation: 存储服务器被配置为从客户端接收存储数据块的请求。 存储数据块的请求由存储服务器通过将数据块压缩成包括多个压缩数据块的压缩组来服务。 存储服务器将压缩组存储在非易失性存储器中，并响应于达到一致性点将压缩组从非易失性存储器刷新到物理存储设备。 通过压缩要存储在存储服务器的系统存储器中的数据，数据存储系统在给定时间段内可以处理的数据量增加。 此外，可以以更低的成本实现性能的提高，因为可以避免额外的物理系统存储器模块的成本，

公开(公告)号：US07774327B1

公开(公告)日：2010-08-10

申请号：US11345249

申请日：2006-01-31

Applicant: Sriram S. Rao ,  Sandeep Yadav ,  John Z. Zhuge

Inventor： Sriram S. Rao ,  Sandeep Yadav ,  John Z. Zhuge

IPC: G06F17/30

CPC classification number: G06F9/4406

Abstract: A storage server maintains an active file system and at least one read-only, persistent point-in-time image (RPPI) of the active file system. The booting time of the storage server is reduced by delaying the mounting of the RPPI(s) until the storage server receives a request to access a corresponding RPPI after the storage server has already booted.

Abstract translation: 存储服务器维护活动文件系统和活动文件系统的至少一个只读持久时间点映像（RPPI）。 存储服务器的启动时间通过延迟RPPI的安装而减少，直到存储服务器在存储服务器已经启动之后接收到访问对应的RPPI的请求。

公开(公告)号：US20100174879A1

公开(公告)日：2010-07-08

申请号：US12349423

申请日：2009-01-06

Applicant: Dnyaneshwar Pawar ,  Rahul Khona ,  Subramanian Periyagaram ,  Sandeep Yadav

Inventor： Dnyaneshwar Pawar ,  Rahul Khona ,  Subramanian Periyagaram ,  Sandeep Yadav

IPC: G06F12/00 ,  G06F12/02 ,  G06F12/16 ,  G06F17/30

CPC classification number: G06F3/0608 ,  G06F3/0611 ,  G06F3/0631 ,  G06F3/064 ,  G06F3/067

Abstract: A method and system for estimating space in a compressed volume to enable a storage server to respond to write requests before actually compressing and/or allocating data on disk. In some embodiments, in response to receiving a request to store data, the storage server estimates the amount of storage space required to store the data on disk. The storage server compares the estimated amount with the amount of available disk space. When the amount of available disk space is less than the estimated space, the storage server sends a response indicating that the request failed. Otherwise, when the amount of available disk space is greater than or equal to the estimate space, the storage server sends a response indicating that the request succeeded. The response is sent before the storage server allocates any disk space in connection with the request.

Abstract translation: 一种用于估计压缩卷空间的方法和系统，以使得存储服务器能够在实际压缩和/或分配磁盘上的数据之前对写请求做出响应。 在一些实施例中，响应于接收到存储数据的请求，存储服务器估计将数据存储在盘上所需的存储空间量。 存储服务器将估计的数量与可用磁盘空间的数量进行比较。 当可用磁盘空间量小于估计的空间时，存储服务器发送一个响应，指示请求失败。 否则，当可用磁盘空间量大于或等于估计空间时，存储服务器发送一个响应，指示该请求成功。 在存储服务器分配与请求相关联的任何磁盘空间之前发送响应。

公开(公告)号：US20100088296A1

公开(公告)日：2010-04-08

申请号：US12245669

申请日：2008-10-03

Applicant: Subramanian Periyagaram ,  Rahul Khona ,  Dnyaneshwar Pawar ,  Sandeep Yadav

Inventor： Subramanian Periyagaram ,  Rahul Khona ,  Dnyaneshwar Pawar ,  Sandeep Yadav

IPC: G06F7/06 ,  G06F17/30

CPC classification number: G06F16/1752 ,  G06F3/061 ,  G06F3/0641 ,  G06F3/067 ,  G06F11/1453 ,  G06F16/183 ,  H04L67/1097 ,  H04L67/2842

Abstract: A technique for organizing data to facilitate data deduplication includes dividing a block-based set of data into multiple “chunks”, where the chunk boundaries are independent of the block boundaries (due to the hashing algorithm). Metadata of the data set, such as block pointers for locating the data, are stored in a tree structure that includes multiple levels, each of which includes at least one node. The lowest level of the tree includes multiple nodes that each contain chunk metadata relating to the chunks of the data set. In each node of the lowest level of the buffer tree, the chunk metadata contained therein identifies at least one of the chunks. The chunks (user-level data) are stored in one or more system files that are separate from the buffer tree and not visible to the user.

Abstract translation: 用于组织数据以便于重复数据删除的技术包括将基于块的数据集划分成多个“块”，其中块边界独立于块边界（由于散列算法）。 数据集的元数据（例如用于定位数据的块指针）被存储在包括多个级别的树结构中，每个级别包括至少一个节点。 树的最低级别包括多个节点，每个节点包含与数据集块大小相关的块元数据。 在缓冲树的最低级别的每个节点中，其中包含的块元数据标识至少一个块。 块（用户级数据）存储在与缓冲树分开的一个或多个系统文件中，并且对用户不可见。

公开(公告)号：US09043555B1

公开(公告)日：2015-05-26

申请号：US12392994

申请日：2009-02-25

Applicant: Rahul Khona ,  Subramaniam Periyagaram ,  Sandeep Yadav ,  Dnyaneshwar Pawar

Inventor： Rahul Khona ,  Subramaniam Periyagaram ,  Sandeep Yadav ,  Dnyaneshwar Pawar

IPC: G06F12/00 ,  G06F3/06

CPC classification number: G06F3/0641 ,  G06F3/0608 ,  G06F3/0683 ,  G06F12/0871 ,  G06F2212/1016 ,  G06F2212/262 ,  G06F2212/263 ,  G06F2212/312 ,  G06F2212/463

Abstract: Provided is a method and system for reducing duplicate buffers in buffer cache associated with a storage device. Reducing buffer duplication in a buffer cache includes accessing a file reference pointer associated with a file in a deduplicated filesystem when attempting to load a requested data block from the file into the buffer cache. To determine if the requested data block is already in the buffer cache, aspects of the invention compare a fingerprint that identifies the requested data block against one or more fingerprints identifying a corresponding one or more sharable data blocks in the buffer cache. A match between the fingerprint of the requested data block and the fingerprint from a sharable data block in the buffer cache indicates that the requested data block is already loaded in buffer cache. The sharable data block in buffer cache is used instead thereby reducing buffer duplication in the buffer cache.

Abstract translation: 提供了一种用于减少与存储设备相关联的缓冲器高速缓存中的重复缓冲器的方法和系统。 减少缓冲区高速缓存中的重复数据包括在尝试将所请求的数据块从文件加载到缓冲区高速缓存中时访问与重复数据删除的文件系统中的文件相关联的文件引用指针。 为了确定所请求的数据块是否已经在缓冲区高速缓存中，本发明的各方面将识别所请求的数据块的指纹与标识缓冲区高速缓存中相应的一个或多个可共享数据块的一个或多个指纹进行比较。 所请求的数据块的指纹与来自缓冲器高速缓存器中的可共享数据块的指纹之间的匹配指示所请求的数据块已经被加载到缓冲器高速缓存中。 使用缓冲区高速缓存中的可共享数据块，从而减少缓冲区高速缓存中的缓冲区重复。

公开(公告)号：US09032527B2

公开(公告)日：2015-05-12

申请号：US13438567

申请日：2012-04-03

Applicant: Pratyusa Kumar Manadhata ,  Sandeep Yadav ,  Prasad V. Rao ,  Marc R. Eisenbarth ,  William G. Horne ,  Stuart Haber

Inventor： Pratyusa Kumar Manadhata ,  Sandeep Yadav ,  Prasad V. Rao ,  Marc R. Eisenbarth ,  William G. Horne ,  Stuart Haber

IPC: G06F12/16 ,  H04L29/06 ,  G06F21/56

CPC classification number: H04L63/145 ,  G06F21/56 ,  H04L61/103 ,  H04L61/2015 ,  H04L63/1433 ,  H04L63/1483 ,  H04L67/02

Abstract: Systems, computer-readable media storing instructions, and methods can infer a state of behavior. Such a method can include constructing a graph including nodes representing hosts and domains based on an event dataset. The graph can be seeded with information external to the event dataset. A belief whether each of the nodes is in a particular state of behavior can be calculated based on marginal probability estimation.

Abstract translation: 系统，存储指令的计算机可读介质和方法可以推断行为状态。 这种方法可以包括基于事件数据集来构建包括表示主机和域的节点的图。 该图可以使用事件数据集外部的信息种子。 可以基于边际概率估计来计算每个节点是否处于特定状态。

公开(公告)号：US08898117B2

公开(公告)日：2014-11-25

申请号：US13589763

申请日：2012-08-20

Applicant: Jim Voll ,  Sandeep Yadav

Inventor： Jim Voll ,  Sandeep Yadav

IPC: G06F17/30 ,  H03M7/30 ,  G06F3/06

CPC classification number: H03M7/30 ,  G06F3/0608 ,  G06F3/0641 ,  G06F3/0643 ,  G06F3/0689

Abstract: A system and method for transparently compressing file system data using compression group descriptors is provided. When data contained within a compression group be compressed beyond a predefined threshold value, a compression group descriptor is included in the compression group that signifies that the data for the group of level 0 blocks is compressed into a lesser number of physical data blocks. When performing a read operation, the file system first determines the appropriate compression group that contains the desired data and determines whether the compression group has been compressed. If so, the file system decompresses the data in the compression group before returning the decompressed data. If the magic value is not the first pointer position, then the data within the compression group was previously stored in an uncompressed format, and the data may be returned without performing a decompression operation.

Abstract translation: 提供了一种使用压缩组描述符对文件系统数据进行透明压缩的系统和方法。 当包含在压缩组中的数据被压缩超过预定义的阈值时，压缩组中包括压缩组描述符，该压缩组表示用于组0级块的数据被压缩成较少数量的物理数据块。 执行读取操作时，文件系统首先确定包含所需数据的适当的压缩组，并确定压缩组是否被压缩。 如果是这样，文件系统在返回解压缩数据之前解压缩压缩组中的数据。 如果魔术值不是第一个指针位置，则压缩组中的数据先前以未压缩格式存储，并且可以在不执行解压缩操作的情况下返回数据。

公开(公告)号：US08892527B1

公开(公告)日：2014-11-18

申请号：US13620684

申请日：2012-09-14

Applicant: Sandeep Yadav ,  Subramanian Periyagaram

Inventor： Sandeep Yadav ,  Subramanian Periyagaram

IPC: G06F7/00 ,  G06F17/30 ,  G06F3/06

CPC classification number: G06F3/0641 ,  G06F3/0608 ,  G06F3/0619 ,  G06F3/067 ,  G06F3/0679 ,  G06F11/1453 ,  G06F12/0871 ,  G06F17/30156 ,  G06F17/30197 ,  G06F17/30386 ,  G06F2206/1014 ,  G06F2212/222

Abstract: A method and system for eliminating the redundant allocation and deallocation of special data on disk, wherein the redundant allocation and deallocation of special data on disk is eliminated by providing an innovate technique for specially allocating special data of a storage system. Specially allocated data is data that is pre-allocated on disk and stored in memory of the storage system. “Special data” may include any pre-decided data, one or more portions of data that exceed a pre-defined sharing threshold, and/or one or more portions of data that have been identified by a user as special. For example, in some embodiments, a zero-filled data block is specially allocated by a storage system. As another example, in some embodiments, a data block whose contents correspond to a particular type document header is specially allocated.

Abstract translation: 一种消除磁盘上特殊数据的冗余分配和释放的方法和系统，其中通过提供专门分配存储系统的特殊数据的创新技术，消除了磁盘上特殊数据的冗余分配和释放。 特别分配的数据是在磁盘上预先分配并存储在存储系统的存储器中的数据。 “特殊数据”可以包括任何预先决定的数据，超过预定义共享阈值的数据的一个或多个部分，和/或用户已经被特别识别的数据的一个或多个部分。 例如，在一些实施例中，零填充数据块由存储系统专门分配。 作为另一示例，在一些实施例中，特别分配其内容对应于特定类型文档头的数据块。