International Business Machines Corporation

James Brian Cunningham

Omar Cardona

Baltazar De Leon, Iii

Matthew Ryan Ochs

Charles Scott Graham

Jeffrey Paul Messing

Herman Dietrich Dierks

Rakesh Sharma

Jorge Rafael Nogueras

Colin Robert DeVilbiss

Jiandi A An

Daniel Gene Eisenhauer

Venkat Venkatsubra

Kyle Alan Lucke

Binh K Hua

Sivarama K Kodukula

Jeffrey James Lynch

Andrew T Koch

Octavian Florin Herescu

Nicholas E Bofferding

Kristopher Craig Whitney

Kiet Hien Lam

Brian James King

Shouchun Li

Baltazar Iii De Leon

Various systems, processes, and products may be used to filter multicast messages in virtual environments. In one implementation, a multicast filtering address is received by a network adapter from at least one of a number of virtual machines. A priority of the multicast filtering address is determined and, based on the priority, the multicast filtering address is stored in either a multicast filtering store of the network adapter or a local filtering store of the at least one virtual machine.

Multicast message filtering in virtual environments

Various systems, processes, and products may be used to filter multicast messages in virtual environments. In one implementation, a multicast filtering address is received by a network adapter. A frequency of use of the multicast filtering address is determined and, based on the frequency of use of the multicast filtering address, the multicast filtering address is stored in either a multicast filtering store of the network adapter or a local filtering store of a respective virtual machine.

Various systems, processes, and products may be used to filter multicast messages in virtual environments. In one implementation, a multicast filtering address is received by a network adapter from at least one of a number of virtual machines. An amount of filtering data is determined corresponding to the at least one virtual machine and, based on the amount of the filtering data corresponding to the at least one virtual machine, the multicast filtering address is stored in either a multicast filtering store of the network adapter or a local filtering store of the at least one virtual machine.

Various systems, processes, and products may be used to filter multicast messages in virtual environments. In one implementation, a multicast filtering address is received by a network adapter from at least one of a number of virtual machines of a computer system. Responsive to receiving the multicast filtering address, a determination is made whether a multicast filtering store of the network adapter is full. Responsive to determining that the multicast filtering store of the network adapter is full, the multicast filtering address is stored in a local filtering store of the at least one virtual machine.

Various systems, processes, and products may be used to filter multicast messages in virtual environments. In one implementation, hardware resources are virtualized to provide a plurality of virtual machines where a number of the virtual machines are configured to receive multicast messages. A network adapter is configured to receive a multicast filtering address from at least one of the number of virtual machines and hash the multicast filtering address to create a hash value. The hash value is linked to the virtual machine via a memory entry.

The present invention provides computer implemented method, system, and computer program product of switching servers without interrupting a client command-response queue. In an embodiment, the present invention includes selecting standby computer server systems, connecting a CRQ of a client computer system to a previous active computer server system and to the selected standby computer server systems, transmitting setup commands to the selected standby computer server systems, communicating data with the previous active computer server system via at least one sub-CRQ, communicating management commands with the previous active computer server system and the selected standby computer server systems via the CRQ, receiving a notification that a switchover has occurred, redirecting the CRQ to one of the selected standby computer server systems, in response to the redirecting, communicating data with a new active computer server system via at least one sub-CRQ, and sending new setup commands to the new computer server system.

Switching servers without interrupting a client command-response queue

Various systems, processes, and products may be used to filter multicast messages in virtual environments. In particular implementations, a system, process, and product for filtering multicast messages in virtual environments may include the ability to, responsive to receiving a multicast message including a destination address, examine a local filtering store of the network adapter for a multicast filtering address matching the destination address. The message is sent to respective virtual machines if the destination address matches the multicast filtering address found in the local filtering store for the respective virtual machines. Responsive to determining that the destination address does not match any multicast filtering address found in the local filtering store, a determination is made whether the local filtering store is full and, if so, the multicast message is dropped.

Various systems, processes, and products may be used to filter multicast messages in virtual environments. In particular implementations, a system, process, and product for filtering multicast messages in virtual environments may include the ability to determine whether a multicast message has been received, examine a local filtering store for a match for the destination address of the multicast message, and send the message to a number of virtual machines if a match for the address is found. The system, process, and product may also include the ability to analyze a filtering store of at least one of the virtual machines for a match for the address if a match is not found in the local filtering store and send the message to a number of the virtual machines if a match for the address is found in a filtering store of one of the virtual machines.

Mechanisms are provided, in a data processing system comprising a host system and a network adapter, for processing received frames of data over a network connection. The mechanisms receive, in the host system from the network adapter, a plurality of frames of data. The mechanisms record, by the host system, for each frame in the plurality of frames, a header size associated with the frame over a current predetermined interval. The mechanisms determine, by the host system, a receive buffer address offset for receive buffers in the host system for a next predetermined interval based on the recorded header sizes of the plurality of frames over the current predetermined interval. In addition, the mechanisms configure, by the host system, the network adapter to utilize the receive buffer address offset to perform data transfers with the host system.

Memory transfer optimization of network adapter data placement when performing header-data split operations

In link aggregation configurations, a data packet may be copied into a buffer space of a first NIC. Load balancing techniques may determine that the packet should be transmitted by a second NIC. The packet exists in memory that the second NIC cannot access. The data packet is copied into memory accessible to the second NIC or the memory location of the packet is registered with the NIC. A copy penalty is incurred if a packet is copied from a first buffer space to a second buffer space. A registration penalty is incurred if the location within the first buffer space is registered with the second NIC. Functionality can be implemented within a link aggregation configuration to register buffer space shared among interconnected NICs. Sharing of buffer space between interconnected NICs allows any one of the NICs to access data within the shared buffer space without incurring a penalty.

Sharing buffer space in link aggregation configurations

A mechanism is provided for providing resource affinity for multi-queue network adapters via dynamic reconfiguration. A device driver allocates an initial queue pair within a memory. The device driver determines whether workload of the data processing system has risen above a predetermined high threshold. Responsive to the workload rising above the predetermined high threshold, the device driver allocates and initializes an additional queue pair in the memory. The device driver programs a receive side scaling (RSS) mechanism in a network adapter to allow for dynamic insertion of an additional processing engine associated with the additional queue pair. The device driver enables transmit tuple hashing to the additional queue pair.

Growing community of inventors

Austin, TX, United States of America

James Brian Cunningham