Cloud Technology Manages Network Bandwidth




email

IBM recently issued a press release touting U.S. Patent No. 8,352,953, titled Dynamically provisioning virtual machines, which was actually issued in January 2013. That suggests that the technology at play in this patent application is making its way into the stream of commerce. This gives us an opportunity to take a look at a real world useful innovation in the cloud computing space and wonder whether the claims issued would satisfy the half of the Federal Circuit that seems opposed to software patents.

The invention at issue was developed as a method for dynamically managing network bandwidth within a cloud computing environment. IBM says this innovation could lead to significant improvements in overall system performance, efficiency and economy.

In a typical cloud computing environment, each user is given access to a virtual machine (“VM”) that delivers a host operating system and physical resources such as processor and memory to support the user’s application requirements. To accommodate numerous users, multiple VMs are assigned within the cloud and as demands for system resources increase and multiply, applications can become constrained by limits on networking bandwidth. This IBM invention allows the system to automatically and dynamically reassign work from one system node to another based on networking bandwidth requirements and availability, ensuring that the system and VMs can run efficiently.

While there are many approaches to enabling virtualization in a cloud, most focus on the issues of CPU and memory utilization and optimization. But even if those functions are managed effectively, overall system performance can be impeded by restricted network bandwidth. By focusing on the networking aspect of system optimization and virtualization, this invention removes a roadblock to overall system efficiency, allowing processing, memory and networking all to work at optimal levels.

Claim 1, the broadest claim found in the ‘953 patent, covers:

A method, comprising: managing a first plurality of virtual machines on a first compute node and a second plurality of virtual machines on a second compute node, wherein the first compute node is coupled to a first Ethernet link of a network switch and the second compute node is coupled to a second Ethernet link of the network switch, and wherein each of the virtual machines is assigned an Internet Protocol address; obtaining data from a management information database of the network switch to determine the amount of network bandwidth being utilized through each of the first and second Ethernet links and to determine the amount of network bandwidth being utilized by the Internet Protocol addresses attributable to each of the virtual machines; and migrating one or more of the virtual machines from the first compute node to the second compute node in response to the first Ethernet link utilizing a greater amount of network bandwidth than the second Ethernet link.

This cloud patent also utilizes a claiming technique whereby the invention is characterized as a non-transitory computer-usable storage medium. Claim 8 recites:

A computer program product including computer-usable program code embodied on a non-transitory computer-usable storage medium, the computer program product including: computer-usable program code for managing a first plurality of virtual machines on a first compute node and a second plurality of virtual machines on a second compute node, wherein the first compute node is coupled to a first Ethernet link of a network switch and the second compute node is coupled to a second Ethernet link of the network switch, and wherein each of the virtual machines is assigned an Internet Protocol address; computer-usable program code for obtaining data from a management information database of the network switch to determine the amount of network bandwidth being utilized through each of the first and second Ethernet links and to determine the amount of network bandwidth being utilized by the Internet Protocol addresses attributable to each of the virtual machines; and computer-usable program code for migrating one or more of the virtual machines from the first compute node to the second compute node in response to the first Ethernet link utilizing a greater amount of network bandwidth than the second Ethernet link.

Claims like these, and many, many others have been drawn into question by the plurality opinion in CLS Bank v. Alice Corporation. Judge Lourie, writing for the plurality, would have determined that the method, computer-readable medium, and system claims are all patent-ineligible subject matter under 35 U.S.C. § 101. This opinion by Judge Lourie is being applied by some panels of the Federal Circuit as precedential, even though there were an equal number of judges who refused to sign on to the decision.

When you look at patent claims issued in a variety of important, useful technologies, you see the real mischief of a decision that renders all claiming techniques for computer software patent ineligible. Eventually, the Supreme Court will need to step in and settle the matter. Given that software is referenced as patentable subject matter throughout the Patent Act, it stretches the imagination that the Supreme Court will ever rule that software is patent ineligible. In the meantime, we wait, wonder and receive decisions that are completely irreconcilable from one another, depending upon the make-up of the Federal Circuit panel.

Tags: , , , , ,

Leave a Reply

You share in the PLI Practice Center community, so we just ask that you keep things civil. Leave out the personal attacks. Do not use profanity, ethnic or racial slurs, or take shots at anyone's sexual orientation or religion. If you can't be nice, we reserve the right to remove your material and ban users who violate our Terms of Service.

You must be logged in to post a comment.