The Path to Grid-Friendly Software Licensing

By By Brooklin J. Gore, Contributing Editor

March 21, 2005

** The views and opinions expressed herein are strictly those of the author and do not necessarily reflect those of Micron Technology Inc. or its subsidiaries or affiliates

Introduction

Grid computing, coupled with companies' ongoing market pressure to reduce operating costs, is changing the way IT applications are run. However, current software licensing methods may restrict the ability to leverage applications with the flexibility that the Grid can provide. This article reviews several licensing schemes in terms of their Grid-friendliness and suggests minor changes to licensing models to better promote Grid usage.

A Review of Licensing Models

Three fundamental licensing models that warrant coverage in any discussion of Grid computing are: 1) user-based, 2) host-based and 3) concurrent.

User-based licensing associates a license with a named user. Typically, named users can upgrade their systems and re-install the software without further involvement of the supplier (i.e., nothing special about a particular machine, like its name or ID, enables or disables the software from operating). In some cases, it is acceptable for these users to have such licensed software installed on more than one machine (e.g., work and home, desktop and laptop, etc.) based on the concept that they can only use a single copy at any given time.

The extreme case of this licensing model is that an application could be available on any host (machine), but is only usable by a finite number of licensed users. One can think of this as a “don't care where, but who” scheme. An advantage is that a named user always has access to the application because nobody else can use it. On the other hand, the license is wasted when the named user is away, for example, on vacation.

Host-based licensing has an opposite focus from user-based licensing. One can think of this as a “don't care who, but where” model. Host-based licenses can either be floating (run on any host) or locked (tied to a specific machine by name or ID). An advantage of this approach is that many users can share a license (e.g., in a training lab or on a shared workstation). A disadvantage is that many licenses can go unused.

Concurrent licensing draws on parts of the two previous schemes. One can think of this approach as “don't care who or where, but how many.” Concurrent licensing models generally involve the use of a license server to keep track of how many licenses are available to be checked out. Generally, a cost will be associated with running such a license server. However, this cost is more than offset by the reduced number of licenses required because this scheme promotes maximal sharing of a license among a group of users. While this approach can facilitate “any time, anywhere” access to a minimally sufficient number of licenses, some vendors artificially limit licenses to a set of nodes or geographical region (i.e., subnet, etc.). These vendors often will allow wider geographical use for an extra fee.

A Changing Computing Landscape

Not too long ago, computing environments were quite static. The user interface consisted of a personal desktop running client-server software connected to back-end systems located in a basement machine room. Today, the use of laptops is on the rise, and employees are increasingly working at home and on the go. On the back end, big iron typically talked to other big iron. More recently, much of the big iron is being replaced by a lot of little iron, but everything is still statically configured: a named machine gets a bunch of software installed based on its role.

In forward-looking companies, however, Grid computing technologies and the continual focus on reducing costs have driven hardware virtualization and consolidation efforts to support greater application portability, reliability and scalability. Many companies are migrating much of their computing infrastructure to a small number of super data centers; maybe one per continent. The transaction that used to run in the basement may now run one or more time zones away.

In some cases, companies may choose to run their applications on someone else's computers. Sun Microsystems recently announced a program to provide CPU cycles by the hour. The bottom line is that the day is rapidly approaching when applications — both user-facing and back-office — will run anywhere. The location and association between an application and the hardware on which it runs will be very dynamic. One may run an application from a box in Austin, Texas, on Monday and the same application from a box in Singapore on Thursday.

Grid-Friendly Software Licensing

While today's Grid-enabled infrastructure supports most of what is discussed above, we're still a few years from seeing widespread availability of Grid-enabled software as the industry sorts out all the standards and value propositions. What is Grid-enabled software? Grid-enabled software actively discovers and utilizes a Grid to improve efficiency, reliability and scalability. As the industry goes along the path to Grid-enabled software, perhaps it should consider an interim step to Grid-friendly software.

The goal of Grid-friendly software is simple: don't do anything to interfere with a Grid execution model. While there are a few technical considerations for Grid-friendly software, the focus of this article is on licensing. Consider how the three licensing models can become more Grid-friendly.

The Grid supports location- (i.e., machine) independent application execution. User-based licensing models could better support this flexibility through a replicated or centralized table of authorized users. Users should be able to run a local copy of “their” software in the morning (maybe a lightweight trial run) and a remote copy via the Grid in the afternoon (a CPU-intensive run that has been matched to a beefy machine, for example). The FlexLM license management system from Macrovision is one tool that supports such an execution model.

The Grid also supports the matching of jobs to machines based on job requirements and machine capabilities. Most Grids can publish arbitrary machine capabilities such as a list of available software. Host-based licensing models can be effectively leveraged on the Grid by such an automated discovery. For example, if a company has purchased 1,000 copies of an application, and 450 are actively being used, the Grid can automatically process a queue of work with the remaining 550 copies. Application providers that use host-based licensing could explicitly state that this use model is acceptable in their licensing agreements.

Perhaps the greatest power of Grids lies in their ability to easily run many jobs in parallel. Concurrent licensing models readily support Grid execution because they generally use a centralized license server. However, usage-based (utility) models would support Grid execution more flexibly and provide better license utilization. For example, if a company was licensed for the concurrent use of 10 copies of software, it could obviously run 10 instances 24 hours in a day for a total of 240 instance-hours. But, on another day, it should be able to run 24 instances for 10 hours. This is also the best model for externally-hosted applications that are run via an Application Service Provider (ASP). Interestingly, the FlexLM license management system mentioned above supports such a licensing model today.

Other Considerations

Please don't make price an issue. Some vendors might want to charge more for the ability to use their application in a Grid. To the extent that customers are just using an off-the-shelf application on the Grid, charging more is not justified. The customers are simply leveraging the investment in their IT infrastructure more wisely. In some cases, when specific functionality is built into the software to explicitly leverage Grid capabilities (like automatically parallelizing tasks perhaps), a price uplift might be justified — but now we're talking Grid-enabled software, not just Grid-friendly software.

It's no longer about location, location, location. From the section on the changing computing landscape, it should be clear that applications increasingly are run independently from user and data location. Licensing schemes that restrict (or charge extra for) use based on geography are antiquated and artificially restrict the way companies must do business to compete in today's global market.

While the thrust of this article is directed at software suppliers, the Grid community is not without involvement. The ability to match jobs to computing resources should be enhanced to support matching on arbitrary resource constraints. Today, the Grid throttles the number of active jobs based on the number of available machines to run those jobs. It would be nice to be able to throttle the active jobs based on some other constraint such as the number of available licenses or the number of available database connections. One Grid computing software vendor, Platform Computing, offers a product called License Scheduler that can manage job deployment based on license resource availability.

Summary

We are entering a new era in location-independent computing flexibility brought forth by Grid technology and driven by a business focus on reduced cost and increased efficiency. While Grid technologies are fairly mature in their ability to match application requirements to compute nodes, perform data transfer, and start and manage remote applications, the applications themselves are not very Grid-friendly. This article suggests the most expeditious path to fully-Grid-enabled applications is for application vendors to first address Grid-unfriendly licensing models and then address the associated technical issues.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion XL — were added to the benchmark suite as MLPerf continues Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire