公开(公告)号：US09699061B2

公开(公告)日：2017-07-04

申请号：US15006848

申请日：2016-01-26

Applicant: Citrix Systems, Inc.

Inventor： Dhiraj Gedam ,  Raveendra Holla ,  Sharvari Mithyantha

IPC: G06F15/16 ,  H04L12/26 ,  H04L12/24

CPC classification number: H04L43/16 ,  H04L41/0213 ,  H04L41/0609 ,  H04L41/0681 ,  H04L43/10

Abstract: In a multi-core device or clustered system, instead of snmpd polling for configured monitoring values of an entity to determine if reached a threshold, each core in a multi-core system or node in a clustered system triggers information to the snmpd about entities that may be or are generating SNMP traps. A configured threshold T is distributed among the cores or nodes, as the case may be, based on the number of cores or nodes. If there are ‘n’ cores in a multi-core device, and the configured threshold is ‘T’, then each core checks for a per-core threshold value ‘T/n’. If there are ‘n’ nodes in a clustered system, and the configured threshold is ‘T’, then each node checks for a per-node threshold value ‘T/n’. Snmpd then gathers information about this entity from all the cores and checks for the threshold ‘T’.

公开(公告)号：US10476764B2

公开(公告)日：2019-11-12

申请号：US14463230

申请日：2014-08-19

Applicant: Citrix Systems, Inc.

Inventor： Dhiraj Gedam

IPC: H04L29/08 ,  H04L29/06 ,  H04L12/26 ,  H04L29/12 ,  H04L29/14

Abstract: The present invention is related to a method for high volume logging for large scale network address translation. A first device intermediary to a plurality of clients and a plurality of database servers allocates a portion of memory to each packet engine in a plurality of packet engines executing on a respective core of a plurality of cores of the first device. The first device establishes large scale network address translation (LSN) for the plurality of clients, the first device logging LSN information of sessions to a corresponding logging buffer established in a respective packet engine's portion of memory. The first device identifies, for a LSN session, a packet engine from the plurality of packet engines to log the information for the LSN session and stores information of the LSN session to the logging buffer in the packet engine's portion of memory.

公开(公告)号：US10015039B2

公开(公告)日：2018-07-03

申请号：US14689933

申请日：2015-04-17

Applicant: Citrix Systems, Inc.

Inventor： Saravana Annamalaisami ,  Raveendra Holla ,  Nishant Kumar Jain ,  Sharvari Mithyantha ,  Dhiraj Gedam

IPC: G06F15/177 ,  H04L12/28 ,  H04L12/24 ,  H04L29/08

CPC classification number: H04L41/0213 ,  H04L41/042 ,  H04L41/046 ,  H04L67/10 ,  H04L67/288

Abstract: The present disclosure is directed towards systems and methods for supporting Simple Network Management Protocol (SNMP) request operations over clustered networking devices. The system includes a cluster that includes a plurality of intermediary devices and an SNMP agent executing on a first intermediary device of the plurality of intermediary devices. The SNMP agent receives an SNMP GETNEXT request for an entity. Responsive to receipt of the SNMP GETNEXT request, the SNMP agent requests a next entity from each intermediary device of the plurality of intermediary devices of the cluster. To respond to the SNMP request, the SNMP agent selects a lexicographically minimum entity. The SNMP agent may select the lexicographically minimum entity from a plurality of next entities received via responses from each intermediary device of the plurality of intermediary devices.

公开(公告)号：US20130275582A1

公开(公告)日：2013-10-17

申请号：US13861702

申请日：2013-04-12

Applicant: CITRIX SYSTEMS, INC.

Inventor： Dhiraj Gedam ,  Raveendra Holla ,  Sharvari Mithyantha

IPC: H04L12/26

CPC classification number: H04L43/16 ,  H04L41/0213 ,  H04L41/0609 ,  H04L41/0681 ,  H04L43/10

Abstract: In a multi-core device or clustered system, instead of snmpd polling for configured monitoring values of an entity to determine if reached a threshold, each core in a multi-core system or node in a clustered system triggers information to the snmpd about entities that may be or are generating SNMP traps. A configured threshold T is distributed among the cores or nodes, as the case may be, based on the number of cores or nodes. If there are ‘n’ cores in a multi-core device, and the configured threshold is ‘T’, then each core checks for a per-core threshold value ‘T/n’. If there are ‘n’ nodes in a clustered system, and the configured threshold is ‘T’, then each node checks for a per-node threshold value ‘T/n’. According to the pigeonhole principle, if an entity has reached or exceeded the threshold ‘T’, then the entity must have reached or exceeded a value of ‘T/n’ on at least one core or node. Upon the entity crossing a ‘T/n’ value on any core or node, the core or node informs snmpd about this entity. Snmpd then gathers information about this entity from all the cores and checks for the threshold ‘T’.

Abstract translation: 在多核设备或集群系统中，代替对配置的实体监视值的snmpd轮询来确定是否达到阈值，集群系统中的多核系统或节点中的每个核心会向snmpd发送有关实体的信息， 可能正在或正在生成SNMP陷阱。 基于核或节点的数量，配置的阈值T分布在核心或节点之间，视情况而定。 如果多核设备中存在“n”个核心，并且配置的阈值为“T”，则每个内核将检查每个核心阈值“T / n”。 如果集群系统中存在“n”个节点，并且配置的阈值为“T”，则每个节点将检查每个节点的阈值“T / n”。 根据排队原则，如果某个实体达到或超过阈值“T”，则该实体必须在至少一个核心或节点上达到或超过“T / n”值。 在实体跨越任何核心或节点上的“T / n”值时，核心或节点通知snmpd关于该实体。 Snmpd然后从所有核心收集有关该实体的信息，并检查阈值“T”。

公开(公告)号：US20160142279A1

公开(公告)日：2016-05-19

申请号：US15006848

申请日：2016-01-26

Applicant: Citrix Systems, Inc.

Inventor： Dhiraj Gedam ,  Raveendra Holla ,  Sharvari Mithyantha

IPC: H04L12/26

CPC classification number: H04L43/16 ,  H04L41/0213 ,  H04L41/0609 ,  H04L41/0681 ,  H04L43/10

Abstract: In a multi-core device or clustered system, instead of snmpd polling for configured monitoring values of an entity to determine if reached a threshold, each core in a multi-core system or node in a clustered system triggers information to the snmpd about entities that may be or are generating SNMP traps. A configured threshold T is distributed among the cores or nodes, as the case may be, based on the number of cores or nodes. If there are ‘n’ cores in a multi-core device, and the configured threshold is ‘T’, then each core checks for a per-core threshold value ‘T/n’. If there are ‘n’ nodes in a clustered system, and the configured threshold is ‘T’, then each node checks for a per-node threshold value ‘T/n’. According to the pigeonhole principle, if an entity has reached or exceeded the threshold ‘T’, then the entity must have reached or exceeded a value of ‘T/n’ on at least one core or node. Upon the entity crossing a ‘T/n’ value on any core or node, the core or node informs snmpd about this entity. Snmpd then gathers information about this entity from all the cores and checks for the threshold ‘T’.

Abstract translation: 在多核设备或集群系统中，代替对配置的实体监视值的snmpd轮询来确定是否达到阈值，集群系统中的多核系统或节点中的每个核心会向snmpd发送有关实体的信息， 可能正在或正在生成SNMP陷阱。 基于核或节点的数量，配置的阈值T分布在核心或节点之间，视情况而定。 如果多核设备中存在“n”个核心，并且配置的阈值为“T”，则每个内核将检查每个核心阈值“T / n”。 如果集群系统中存在“n”个节点，并且配置的阈值为“T”，则每个节点将检查每个节点的阈值“T / n”。 根据排队原则，如果某个实体达到或超过阈值“T”，则该实体必须在至少一个核心或节点上达到或超过“T / n”值。 在实体跨越任何核心或节点上的“T / n”值时，核心或节点通知snmpd关于该实体。 Snmpd然后从所有核心收集有关该实体的信息，并检查阈值“T”。

公开(公告)号：US09294381B2

公开(公告)日：2016-03-22

申请号：US13861702

申请日：2013-04-12

Applicant: Citrix Systems, Inc.

Inventor： Dhiraj Gedam ,  Raveendra Holla ,  Sharvari Mithyantha

IPC: G06F15/16 ,  H04L12/26 ,  H04L12/24

CPC classification number: H04L43/16 ,  H04L41/0213 ,  H04L41/0609 ,  H04L41/0681 ,  H04L43/10

Abstract: In a multi-core device or clustered system, instead of snmpd polling for configured monitoring values of an entity to determine if reached a threshold, each core in a multi-core system or node in a clustered system triggers information to the snmpd about entities that may be or are generating SNMP traps. A configured threshold T is distributed among the cores or nodes, as the case may be, based on the number of cores or nodes. If there are ‘n’ cores in a multi-core device, and the configured threshold is ‘T’, then each core checks for a per-core threshold value ‘T/n’. If there are ‘n’ nodes in a clustered system, and the configured threshold is ‘T’, then each node checks for a per-node threshold value ‘T/n’. According to the pigeonhole principle, if an entity has reached or exceeded the threshold ‘T’, then the entity must have reached or exceeded a value of ‘T/n’ on at least one core or node. Upon the entity crossing a ‘T/n’ value on any core or node, the core or node informs snmpd about this entity. Snmpd then gathers information about this entity from all the cores and checks for the threshold ‘T’.

Abstract translation: 在多核设备或集群系统中，代替对配置的实体监视值的snmpd轮询来确定是否达到阈值，集群系统中的多核系统或节点中的每个核心会向snmpd发送有关实体的信息， 可能正在或正在生成SNMP陷阱。 基于核或节点的数量，配置的阈值T分布在核心或节点之间，视情况而定。 如果多核设备中存在“n”个核心，并且配置的阈值为“T”，则每个内核将检查每个核心阈值“T / n”。 如果集群系统中存在“n”个节点，并且配置的阈值为“T”，则每个节点将检查每个节点的阈值“T / n”。 根据排队原则，如果某个实体达到或超过阈值“T”，则该实体必须在至少一个核心或节点上达到或超过“T / n”值。 在实体跨越任何核心或节点上的“T / n”值时，核心或节点通知snmpd关于该实体。 Snmpd然后从所有核心收集有关该实体的信息，并检查阈值“T”。

公开(公告)号：US20160057031A1

公开(公告)日：2016-02-25

申请号：US14463230

申请日：2014-08-19

Applicant: CITRIX SYSTEMS, INC.

Inventor： Dhiraj Gedam

IPC: H04L12/26 ,  H04L29/12

CPC classification number: H04L43/04 ,  H04L43/08 ,  H04L61/2514 ,  H04L61/2517 ,  H04L61/2535 ,  H04L63/0272 ,  H04L63/0281 ,  H04L67/1034 ,  H04L69/40

Abstract: The present invention is related to a method for high volume logging for large scale network address translation. A first device intermediary to a plurality of clients and a plurality of database servers allocates a portion of memory to each packet engine in a plurality of packet engines executing on a respective core of a plurality of cores of the first device. The first device establishes large scale network address translation (LSN) for the plurality of clients, the first device logging LSN information of sessions to a corresponding logging buffer established in a respective packet engine's portion of memory. The first device identifies, for a LSN session, a packet engine from the plurality of packet engines to log the information for the LSN session and stores information of the LSN session to the logging buffer in the packet engine's portion of memory.

Abstract translation: 本发明涉及用于大规模网络地址转换的大容量日志记录方法。 多个客户机和多个数据库服务器的第一设备中介向多个分组引擎中的每个分组引擎分配一部分存储器，该多个分组引擎在第一设备的多个核心的相应核心上执行。 第一设备为多个客户端建立大规模网络地址转换（LSN），第一设备将会话的LSN信息记录到相应的分组引擎的存储器部分中建立的对应的记录缓冲区。 第一设备为LSN会话识别来自多个分组引擎的分组引擎，以记录LSN会话的信息，并将LSN会话的信息存储在分组引擎的存储器部分中的记录缓冲区。

公开(公告)号：US20150222475A1

公开(公告)日：2015-08-06

申请号：US14689933

申请日：2015-04-17

Applicant: Citrix Systems, Inc.

Inventor： Saravana Annamalaisami ,  Raveendra Holla ,  Nishant Kumar Jain ,  Sharvari Mithyantha ,  Dhiraj Gedam

IPC: H04L12/24

CPC classification number: H04L41/0213 ,  H04L41/042 ,  H04L41/046 ,  H04L67/10 ,  H04L67/288

Abstract: The present disclosure is directed towards systems and methods for supporting Simple Network Management Protocol (SNMP) request operations over clustered networking devices. The system includes a cluster that includes a plurality of intermediary devices and an SNMP agent executing on a first intermediary device of the plurality of intermediary devices. The SNMP agent receives an SNMP GETNEXT request for an entity. Responsive to receipt of the SNMP GETNEXT request, the SNMP agent requests a next entity from each intermediary device of the plurality of intermediary devices of the cluster. To respond to the SNMP request, the SNMP agent selects a lexicographically minimum entity. The SNMP agent may select the lexicographically minimum entity from a plurality of next entities received via responses from each intermediary device of the plurality of intermediary devices.

Abstract translation: 本公开涉及用于支持通过集群网络设备的简单网络管理协议（SNMP）请求操作的系统和方法。 该系统包括一组包括多个中间装置和在多个中间装置的第一中介装置上执行的SNMP代理。 SNMP代理接收一个实体的SNMP GETNEXT请求。 响应于接收到SNMP GETNEXT请求，SNMP代理从集群的多个中间设备的每个中间设备请求下一个实体。 为了响应SNMP请求，SNMP代理选择一个字典最小实体。 SNMP代理可以从通过多个中间设备的每个中间设备的响应接收的多个下一个实体中选择词典上的最小实体。