International Business Machines Corporation

Lenovo (singapore) Pte. Ltd.

Ward Ray Nelson

Josep Cors

Jun C Yin

David Alan Christenson

David R Engebretsen

Paul Edward Movall

Chandramouli Radhakrishnan

Adam Charles Lange-Pearson

David Lee Hermsmeier

Daniel E Pradilla

Gary Dean Anderson

Gerald Kreissig

Andreas Arnez

Joshua W Boyer

Clinton Gene Laschkewitsch

Methods, computer program products, and systems to assist in distribution of packets between a plurality of switch elements in a distributed switch, by monitoring an attribute of each of the plurality of switch elements including a first switch element and a first surrogate switch element, the first surrogate switch element in a first level of a hierarchy and configured using a default hierarchy configuration, the default hierarchy configuration forwarding a data frame to at least one of: a destination switch element of the data frame, and a second surrogate switch element, the second surrogate switch element in a second hierarchy level; computing a score, based on the monitored attribute, for each of the switch elements; and upon determining the score of the first switch element is greater than the score of the first surrogate switch element, designating the first switch element as the first surrogate switch element.

System guided surrogating control in broadcast and multicast

Techniques are provided for providing access control lists in a distributed network switch. The distributed network switch made of switch units is divided into logical switch partitions, or logical networks. Physical ports of the switch units are partitioned into logical ports, where each logical port is associated with a logical switch partition. A control point of the distributed network switch manages and assigns a service tag (S-Tag) used to identify which logical port ingress and egress frames are associated with. To generate metrics and other forwarding actions for a given logical switch partition, the control point sets up access control list (ACLs) targeting the logical port associated with the S-Tags associated with the given logical switch partition.

Metrics and forwarding actions on logical switch partitions in a distributed network switch

Systems and methods to analyze layer-2 data frame switch forwarding are provided. A first switch may be coupled to a second switch. The first switch may maintain state information mirroring the state of the second switch. A sequence number may be appended to a data frame that is received at the first switch. Forwarding compliance of the second switch may be determined by analyzing a runtime attribute of the first switch.

Data link layer switch frame forwarding analysis

A firmware model is provided for a shared port membership table that is partitioned for different needs of a distributed network switch, such as broadcast groups, multicast groups, ACL rules, and other port membership groupings. The shared port membership table enables a control point to set the size of each of the types of port membership information based on their usage or expected usage. By shrinking one partition, the control point is able to use the port group entries for a different partition of port group information.

Port membership table partitioning

The different switch modules making up a distributed virtual switch may route configuration commands for hardware resources to different modules within the distributed switch using a distribution and routing layer. At least one of the switch modules maintains a routing table that defines which switch modules are responsible for which hardware resources. The switch module uses the routing tables to forward the commands on the distribution and routing layer to the responsible switch module which then ensures that the relevant hardware resources are configured.

Distributing functions in a distributed and embedded environment

Method to assist in distribution of packets between a plurality of switch elements in a distributed switch, by monitoring an attribute of each of the plurality of switch elements including a first switch element and a first surrogate switch element, the first surrogate switch element in a first level of a hierarchy and configured using a default hierarchy configuration. The default hierarchy configuration forwards a data frame to at least one of: a destination switch element of the data frame, and a second surrogate switch element. The second surrogate switch element is in a second hierarchy level. A score is computed based on the monitored attribute for each of the switch elements. Upon determining the score of the first switch element is greater than the score of the first surrogate switch element, the first switch element is designated as the first surrogate switch element.

Method, computer program product, and system to assist in distribution of packets between a plurality of switch elements in a distributed switch, by monitoring an attribute of each of the plurality of switch elements including a first switch element and a first surrogate switch element, the first surrogate switch element in a first level of a hierarchy and configured using a default hierarchy configuration, the default hierarchy configuration forwarding a data frame to at least one of: a destination switch element of the data frame, and a second surrogate switch element, the second surrogate switch element in a second hierarchy level; computing a score, based on the monitored attribute, for each of the switch elements; and upon determining the score of the first switch element is greater than the score of the first surrogate switch element, designating the first switch element as the first surrogate switch element.

Distributed event management in an embedded support processor computer system includes an embedded support processor having an event distribution agent used to communicate with local processes internal to the embedded support processor. A process communicates with the event distribution agent of the embedded support processor and registers for one or more events. Another process signals an event to the event distribution agent of the embedded support processor. The event distribution agent of the embedded support processor notifies each registered process of the signaled event. The event distribution agent of the embedded support processor is used to communicate with a peer event distribution agent included in a main processor of the embedded support processor computer system, and to communicate over a network with a peer event distribution agent included in an attached device, such as a personal computer.

Method and apparatus for implementing distributed event management in an embedded support processor computer system

A recovery process for embedded processors monitors other processes in the system. Each process may specify a recovery policy residing in nonvolatile electronic memory that preferably includes a recovery count, a recovery time, and a recovery action. If a process terminates unexpectedly, the recovery process determines whether the process had a corresponding recovery policy. If not, the recovery process does not recover the process. If the process has a corresponding recovery policy, the recovery process determines whether it can recover the process by examining the recovery count and recovery time specified in the recovery policy. If the process can be recovered, the recovery process performs the recovery action specified in the corresponding recovery policy. If the process cannot be recovered, the recovery process resets the system. In this manner, a single process may be recovered by the recovery process without user intervention and without affecting other processes running on the system, and without performing a reset of the entire system.

Growing community of inventors

Stewartville, MN, United States of America

Ward Ray Nelson