News Column

Researchers Submit Patent Application, "Dynamic Optimization of Command Issuance in a Computing Cluster", for Approval

February 20, 2014



By a News Reporter-Staff News Editor at Computer Weekly News -- From Washington, D.C., VerticalNews journalists report that a patent application by the inventors Cudak, Gary D. (Creedmoor, NC); Hardee, Christopher J. (Raleigh, NC); Humes, Randall C. (Raleigh, NC); Roberts, Adam (Moncure, NC); Suffern, Edward S. (Chapel Hill, NC), filed on July 24, 2012, was made available online on February 6, 2014.

The patent's assignee is International Business Machines Corporation.

News editors obtained the following quote from the background information supplied by the inventors: "The present invention relates to cluster based computing and more particularly to command propagation in a computing cluster.

"Computing clusters have become common in the field of high-availability and high-performance computing. Cluster-based systems exhibit three important and fundamental characteristics or properties: reliability, availability and serviceability. Each of these features is of paramount importance when designing the software and the hardware of a new robust clustered system. As opposed to the symmetric multi-processing (SMP) systems whose scalability can be limited and which can result in substantially diminished returns upon the addition of processors to the system, a clustered-based system consists of multiple computers that are connected over high-speed network communicative linkages.

"Each computer in a cluster enjoys its own memory, possibly its own disk space and it hosts its own local operating system. Each node within the cluster system can be viewed as a processor-memory module that cooperates with other nodes such that it can provide system resources and services to user applications. Nodes in a cluster system, however, are not limited to a physical computing system. Rather, nodes in a cluster system also can include virtual machines operating in a physical host environment.

"Clusters can be characterized by increased availability since the failure of a particular node does not affect the operation of the remaining nodes. Rather, any one failed node can be isolated and no longer utilized by the cluster-based system until the node can be repaired and incorporated again within the cluster. Additionally, the load of a failed node within a cluster can be equitably shared among the functional nodes of the cluster. Thus, clusters have proven to be a sensible architecture for deploying applications in the distributed environment and clusters are now the platform of choice in scalable, high-performance computing.

"Individual computing nodes of a cluster can be managed at the command line or programmatically. Generally, commands are issued from a command source within a process execution space and can be directed to one or more targeted nodes. Typical commands include node reboot, update, launch and shut down to name a few possible commands. In some instances, the same command can be directed to multiple different nodes within a cluster.

"When the same command is directed to multiple different nodes within a cluster, the sequence in which the commands are individually transmitted is arbitrary. However, depending upon network conditions, those commands may be received in each of the targeted nodes at different times. In the case of high network latency for some of those nodes, the gap in time of receipt of a command from a node receiving the command soonest and the node receiving the command latest can be intolerable and can result in an unwanted system condition."

As a supplement to the background information on this patent application, VerticalNews correspondents also obtained the inventors' summary information for this patent application: "Embodiments of the present invention address deficiencies of the art in respect to command and control in a computing cluster and provide a novel and non-obvious method, system and computer program product for dynamic optimization of command sequencing in a computing cluster. In an embodiment of the invention, a method for dynamic optimization of command sequencing in a computing cluster can include selecting a set of nodes in the computing cluster to receive a single command and probing the nodes of the set to compute a computer communications latency for each of the nodes. Thereafter, the single command can be transmitted to the nodes in the set according to a sequence determined by the computed computer communications latency for each of the nodes.

"In one aspect of the embodiment, the nodes are probed using a ping directive to each of the nodes. In this instance, the computed computer communications latency accounts for a time consumed in sending data to the nodes using the ping directive. In another aspect of the embodiment, the nodes are probed using a traceroute directive to each of the nodes. In this instance, the computed computer communications latency accounts for a number of hops necessary to reach each of the nodes using the traceroute directive. In yet another aspect of the embodiment, the nodes are probed with a test directive and the computed computer communications latency accounts for a period of time consumed by the nodes to each complete performance of the test directive. In even yet another aspect of the embodiment, the sequence is managed within a data structure of nodes sorted according to the computed computer communications latency for each of the nodes.

"In another embodiment of the invention, a clustered computing system can be configured for dynamic optimization of command sequencing. The system can include one or more different computers, each including memory and at least one processor. Each of the computers further can be coupled to the other over a computer communications network. The system also can include multiple different nodes each supported by one or more of the computers and forming a computing cluster. Finally, the system can include a command sequencer executing in memory of one of the computers in association with an originating one of the nodes. The sequencer can include program code enabled to select a set of the nodes to receive a single command, to probe the nodes of the set to compute a computer communications latency for each of the nodes, and to transmit the single command to the nodes in the set according to a sequence determined by the computed computer communications latency for each of the nodes.

"Additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The aspects of the invention will be realized and attained by means of the elements and combinations particularly pointed out in the appended claims. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

"The accompanying drawings, which are incorporated in and constitute part of this specification, illustrate embodiments of the invention and together with the description, serve to explain the principles of the invention. The embodiments illustrated herein are presently preferred, it being understood, however, that the invention is not limited to the precise arrangements and instrumentalities shown, wherein:

"FIG. 1 is a pictorial illustration of a process for dynamic optimization of command sequencing in a computing cluster;

"FIG. 2 is a schematic illustration of a clustered computing system configured for dynamic optimization of command sequencing; and,

"FIG. 3 is a flow chart illustrating a process for dynamic optimization of command sequencing in a computing cluster."

For additional information on this patent application, see: Cudak, Gary D.; Hardee, Christopher J.; Humes, Randall C.; Roberts, Adam; Suffern, Edward S. Dynamic Optimization of Command Issuance in a Computing Cluster. Filed July 24, 2012 and posted February 6, 2014. Patent URL: http://appft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=%2Fnetahtml%2FPTO%2Fsearch-adv.html&r=760&p=16&f=G&l=50&d=PG01&S1=20140130.PD.&OS=PD/20140130&RS=PD/20140130

Keywords for this news article include: International Business Machines Corporation.

Our reports deliver fact-based news of research and discoveries from around the world. Copyright 2014, NewsRx LLC


For more stories covering the world of technology, please see HispanicBusiness' Tech Channel



Source: Computer Weekly News


Story Tools