International Business Machines Corporation

Gary Roy Shippy

Dhruv Manmohandas Desai

Michael Leo Scollard

Nickolas John Gruendler

Matthew T Belanger

Steven Wade Hunter

Tzvetelina B Battestilli

Malcolm Scott Ware

Christopher Landon Wood

James Edward Hughes

Michael Joseph Steinmetz

Carl A Morrell

Roch A Guerin

Gerald Lebizay

Challis Llewellyn Purrington

Raif O Onvural

Ling-Ching W Tai

Terry Lee Nelms

Caglan M Aras

Terry L Nelms, Ii

Lance Warren Russell

William Gavin Holland

Sumeet Kochar

Thomas M Brey

John T Lloyd

Alfredo Aldereguia

Scott Neil Dunham

Randolph Scot Kolvick

Brahmanand Kumar Gorti

Charles H Jolissaint

Kian-Bon K Sy

Edward Klodnicki

Henry Gaines McMillan

Paul Anthony Smith

Mack Edward Phelps

Lynn Erich Petersen

Joe Wayne Blackard

John G Stevenson

Stephen Anton Owen

Brian E Bigelow

Ronald D Martin

Maya P Yarbrough

Richard Adams Gillaspy

Joesph E Bolan

John Douglas Wright

William C Schaal

Lynn Erich Peterson

William John Henthorn

Steering data communications packets for transparent, bump-in-the-wire data processing among multiple service applications; the service applications executing on servers organized in groups, each server in a server group executing an instance of a same service application, including selecting an instance of each of the service applications to process a packet; identifying a server selection value for each packet as a single integer value that yields server selection modulus values that identify each instance of each of the service applications to process the packet; and steering, by the switches in dependence upon the server selection modulus values, each packet to each instance of each of the service applications to process the packet.

Steering data communications packets among service applications with server selection modulus values

Link-level data communications implemented in switching apparatus comprising modular switches disposed within a modular computer cabinet that includes modular computer systems; the switching apparatus configured as two layers of switches, the first layer switches coupled to one another for communications by inter-switch links, each second layer switch coupled for communications to the modular computer systems; all the switches stacked by a stacking protocol that shares administrative configuration information among the switches through the inter-switch links and presents all the switches as a single logical switch; the switching apparatus including ports coupling the apparatus to networks and to service applications and terminating applications on the modular computer systems; and sending the packet from network to modular computer system to which the packet is directed, or from modular computer system to network to which the packet is directed, the packet traversing none of the inter-switch links among the first layer switches.

Two-layer switch apparatus avoiding first layer inter-switch traffic in steering packets through the apparatus

A method, switch, and/or computer program product routes IP packet flows. An Ethernet switch receives an IP packet flow. Each of the packets in the IP packet flow has a header that contains a same 5-tuple. A load balancing control engine determines whether servers in a group of servers are balanced in their utilization according to 5-tuple redirection rules contained in the load balancing control engine. In response to the load balancing control engine determining, according to the 5-tuple redirection rules, that the servers are balanced, the Ethernet switch routes the IP packet flow to the servers. In response to the load balancing control engine determining that the servers are unbalanced, the load balancing control engine instructs the Ethernet switch to redirect the IP packet flow to a server that is relatively less busy than other servers.

Load-balancing via modulus distribution and TCP flow redirection due to server overload

A method, system, and/or computer program product steers Internet Protocol (IP) packet fragments. A switch receives a first IP packet fragment that contains a 3-tuple in its header. In response to determining that the first IP packet fragment is not a lead packet fragment in the series of IP packet fragments, the first IP packet fragment is pushed onto a data stack. The switch subsequently receives a second IP packet fragment that contains a 5-tuple, which includes the 3-tuple from the first IP packet fragment plus a source port and destination port used by the second IP packet fragment. The 3-tuple is mapped to the 5-tuple in a fragmentation table, thus enabling both the first IP packet fragment and the second IP packet fragment to be steered to the destination port to which the second IP packet fragment is being sent.

Steering fragmented IP packets using 5-tuple based rules

A method, system and/or computer program product route IP packet flows. An Ethernet switch is coupled to a load balancing control engine, which contains load balancing logic that is logically isolated from the Ethernet switch. The Ethernet switch, which routes IP packet flows to servers, receives an IP packet flow. The load balancing control engine determines if the servers are balanced in their workloads. In response to the load balancing control engine determining that the servers are balanced, the Ethernet switch routes the IP packet flow to the servers without the Ethernet switch directly receiving any feedback from the servers regarding their workloads. In response to the load balancing control engine determining that the servers are unbalanced, the load balancing control engine instructs the Ethernet switch to redirect the IP packet flow to a server that is relatively less busy than other servers.

A method, system and/or computer program steer internet protocol (IP) packet fragments that are components of a series of IP packet fragments. A switch receives an IP packet fragment. In response to determining that the fragment is not a lead packet fragment in a series of IP packet fragments that make up an original IP packet, the IP packet fragment is pushed onto a data stack. The switch then receives an IP packet fragment which is determined to be the lead packet fragment in a series of IP packet fragments. The IP 5-tuple from the lead packet fragment is parsed to steer all fragments in the series to a destination port.

Link-level data communications carried out in a link-level data communications switching apparatus that includes modular link-level data communications switches; the switching apparatus is configured as two layers of link-level data communications switches; all the switches stacked by a stacking protocol that shares administrative configuration information among the switches and presents the switches as a single logical switch; the switching apparatus includes data communications ports coupling the switching apparatus to data communications networks and to service applications, each service application associated with a unique, link-level identifier; the switching apparatus includes rules governing the steering of packets among service applications and networks; including receiving, in the switching apparatus, packets directed to a destination network; and steering each packet among the service applications to the destination network in accordance with the rules, without using the link-level identifier of any service application.

Two-layer switch apparatus to avoid first layer inter-switch link data traffic in steering packets through bump-in-the-wire service applications

Steering data communications packets among multiple service applications in a link-level data communications switching apparatus that includes a link-level data communications switch and data communications ports coupling the switching apparatus to networks organized into pairs of networks, and ports connected to service applications, the switching apparatus further including rules governing the steering of data communications among service applications and networks, at least one rule that includes a network code that identifies a network pair and a direction of travel between the networks, including receiving data communications packets directed to a destination network, each packet containing a source network address, and steering by the switching apparatus each packet, the steering carried out only in accordance with the rules, using neither the source network address of the packet, the destination network address of the packet, nor the link-level identifier of any service application.

Source-based steering data communications packets for transparent, bump-in-the-wire processing among multiple service applications

Steering data communications packets for transparent, bump-in-the-wire processing among multiple data processing applications in a link-level data communications switch, the switch including data communications ports, a port connected to a first data communications network and another port connected to a second data communications network, other ports connected to the data processing applications, the switch also including rules governing the steering, including receiving in the switch from the first network data communications packets directed to the second network, each packet containing a source network address and a destination network address and steering by the switch each packet among the applications and through an egress port, such steering carried out only in accordance with the rules, using neither the source network address of the packet, the destination network address of the packet, nor any link-level identifier of any application.

Steering data communications packets for transparent bump-in-the-wire processing among multiple data processing applications

A method and system and calibration technique for power measurement and management over multiple time frames provides responsive power control while meeting global system power consumption and power dissipation limits. Power output of one or more system power supplies is measured and processed to produce power values over multiple differing time frames. The measurements from the differing time frames are used to determine whether or not system power consumption should be adjusted and then one or more devices is power-managed in response to the determination. The determination may compare a set of maximum and/or minimum thresholds to each of the measurements from the differing time frames. A calibration technique uses a precision reference resistor and voltage reference controlled current source to introduce a voltage drop from the input side of a power supply sense resistor calibration is made at the common mode voltage of the power supply output.

Method, system and calibration technique for power measurement and management over multiple time frames

Power is allocated to blades based on an estimate of the actual power they are expected to use rather than their maximum-power draw-value. To protect against situations where the estimated actual-power draw-value is exceeded, a hardware comparator monitors the blade system load against a predetermined threshold value set by a management module (MM) based on user input. If this threshold value is exceeded, a throttle latch is triggered, based on a signal from a service processor monitoring the blade system load. The output of this latch directly engages throttling. The service processor also monitors the output of the latch and communicates information regarding the throttling to the MM for evaluation.

Throttle management for blade system

A calibration technique provides for precision calibration of system power measurement. The calibration technique uses a precision reference resistor and voltage reference controlled current source to introduce a voltage drop from the input side of a power supply sense resistor. Calibration measurements are thereby made for voltages substantially equal to the voltage of the power supply output, so that differential measurements are made at substantially the same common-mode voltage by appropriate selection of the relationship of the resistance of the precision reference resistor and the resistance of the sense resistor, where the common-mode voltage of the measurement is substantially equal to the power supply output voltage.

Calibration technique for power measurement and management over multiple time frames

A method and system are disclosed to enable and control power reduction in a blade/chassis system. A 'maximum power reduction' attribute is stored in the VPD of the blade (or can otherwise be input to or retrieved or calculated by the management entity). The management module of the chassis in which the blades and power supplies are located uses this information to manage the power reduction of blades when the system is operating in an over-subscription mode and a power supply fails. If throttling is required, the system knows the amount of power reduction available for each blade and controls the throttling by spreading it out among the blades in the system so that, ideally, no blade will cease operation altogether.

Programmable throttling in blade/chassis power management

A method and system are disclosed to enable and control over-subscription in a blade/chassis system and to provide the capability to recover in the event of the loss of a redundant power supply. An over-subscription policy is determined by a system administrator and is set in the chassis management module. Information regarding the maximum power allocation needed for each blade being powered by the power supply system is identified by the blade and stored in its VPD or otherwise made available to the chassis management module. The management module of the chassis in which the blades and power supplies are located uses this information to manage the subscription of blades and the over-subscription of blades when appropriate. If throttling is required, the system also allows a predetermined blade priority to be used to identify which blades will be reduced in power.

Power management using spare capacity of redundant power supply in blade environment

Growing community of inventors

Cary, NC, United States of America

Gary Roy Shippy