公开(公告)号：US20070027941A1

公开(公告)日：2007-02-01

申请号：US11191777

申请日：2005-07-27

申请人： Ralph Attila Becker-Szendy ,  Richard Golding ,  Darrell Long

发明人： Ralph Attila Becker-Szendy ,  Richard Golding ,  Darrell Long

IPC分类号： G06F17/30

CPC分类号： G06F9/5011 ,  G06F9/5016

摘要： A resource utilization enforcement system enforces resource quotas in a distributed system. A bank on a central server maintains an account for each principal; the account is equivalent to a resource quota for the principal. Quotas are tracked through the use of digital coins that represent resource consumption. The bank is allowed coin generation privileges. At initiation of a computing session, a purse manager on the client accesses the account of the principal and requests coins to exchange for consumed resources. The purse manager manages the coins withdrawn in a “purse” for the principal. The purse manager submits coins to a cashier on the storage device to “purchase” storage for the principal. Refunds are provided if the storage consumption event frees storage.

摘要翻译： 资源利用执行系统在分布式系统中实施资源配额。 中央服务器上的银行为每个主体维护一个帐户; 该帐户相当于主体的资源配额。 通过使用代表资源消耗的数字硬币来跟踪配额。 银行允许硬币产生特权。 在计算会话开始时，客户端的钱包管理器访问主体的帐户并请求硬币来交换消耗的资源。 钱包经理管理用于委托人的“钱包”中提取的硬币。 钱包经理将硬币提交到存储设备上的收银员以为主体“购买”存储。 如果存储消耗事件释放存储空间，则提供退款。

公开(公告)号：US07962563B2

公开(公告)日：2011-06-14

申请号：US11388524

申请日：2006-03-24

申请人： Ralph Attila Becker-Szendy ,  Richard Andrew Golding ,  Caixue Lin ,  Theodore Ming-Tao Wong ,  Ahmed Zaki Omer

发明人： Ralph Attila Becker-Szendy ,  Richard Andrew Golding ,  Caixue Lin ,  Theodore Ming-Tao Wong ,  Ahmed Zaki Omer

IPC分类号： G06F15/16

CPC分类号： G06F17/30067

摘要： A scheduler selects an I/O from a session of a pool and updates token buckets associated with resource limits and reserves for the session and the pool and statistics used in determining fair sharing. To select an I/O, the scheduler identifies sessions with a non-empty queue, identifies head I/Os in the queues, computes for the head I/O a deadline using session and pool reserve buckets and a release time using session and pool limit buckets, and selects a head I/O with an earliest deadline that is past the release time. If the deadline of the selected candidate head I/O is in the past, the scheduler transfers the selected head I/O to the tail of the storage device queue. Otherwise, the scheduler selects the pool with the least amount of I/O traffic according to a session fair share estimator.

摘要翻译： 调度器从池的会话中选择I / O，并更新与会话的资源限制和预留相关联的令牌桶以及用于确定公平共享的池和统计信息。 为了选择I / O，调度程序将识别具有非空队列的会话，识别队列中的头I / O，使用会话和池预留存储桶计算头I / O截止期限，并使用会话池和池 限制桶，并选择一个超过发布时间的最早期限的头I / O。 如果所选择的候选头I / O的最后期限在过去，调度器将所选择的头I / O传送到存储设备队列的尾部。 否则，调度程序根据会话公平共享估计器选择具有最少I / O流量的池。

公开(公告)号：US07243089B2

公开(公告)日：2007-07-10

申请号：US10723750

申请日：2003-11-25

申请人： Ralph Attila Becker-Szendy ,  Miriam Sivan-Zimet

发明人： Ralph Attila Becker-Szendy ,  Miriam Sivan-Zimet

IPC分类号： G06F17/30 ,  G06F17/00 ,  G06F12/00

CPC分类号： G06F17/30067 ,  Y10S707/99931 ,  Y10S707/99945

摘要： Indirect access to local file systems is provided using storage tank protocols, allowing federation of a local file system into a distributed file system while preserving local access to the existing data in the local file system. The goal of the present system is to federate and migrate the data on a computer system with minimum disruption to applications operating on the computer system. Existing applications on a computer system continue to operate during data federation and migration and require little or no reconfiguration either when the data migration starts or when it ends. Data consistency is maintained: existing applications may modify data in the file system during migration or federation. During federation, other computer systems (or hosts) may modify the data in the file system if access control information allows them to do so. All changes in the file system are seen consistently on all hosts. Minimal downtime is required to install the present system and reconfigure the existing applications to communicate with the present system.

摘要翻译： 使用存储池协议提供对本地文件系统的间接访问，允许将本地文件系统联合到分布式文件系统中，同时保留对本地文件系统中现有数据的本地访问。 本系统的目标是在计算机系统上联合和迁移数据，对计算机系统上运行的应用程序的影响最小。 计算机系统上的现有应用程序在数据联合和迁移期间继续运行，并且在数据迁移开始或结束时都需要很少或不需要重新配置。 保持数据一致性：现有应用程序可能在迁移或联合期间修改文件系统中的数据。 在联盟期间，如果访问控制信息允许他们这样做，其他计算机系统（或主机）可以修改文件系统中的数据。 文件系统中的所有更改都会在所有主机上一致显示。 安装本系统需要最少的停机时间，并重新配置现有应用程序以与本系统进行通信。