The Grid-Cloud Connection (Pt. I): Compare and Contrast

By Derrick Harris

October 8, 2008

Grid computing. Cloud computing. Are there any IT paradigms that have garnered more hype and more skepticism without most people even knowing what they mean? Probably not, but maybe that is because the terms themselves have no real meanings to most IT consumers, just connotations.

And connotations can be scary. Burned to some degree by the existing confusion surrounding grid computing, many grid vendors have drastically cut the term from their marketing strategies. Learning from what might be perceived as mistakes, these vendors are not so quick to latch onto cloud computing. However, many of their new directions could easily fall under the cloud umbrella, and those in the know readily acknowledge that grid technologies underlie the cloud.

So, what’s a middleware vendor to do?

First, Compare

Within the Data Center Business Division at Univa UD, messaging around grid computing has been all but eliminated as the division attempts to build traction for its Reliance datacenter orchestration product (from which the company also has nixed the Grid MP middleware component). What division general manager Alex Brown calls the “traffic cop or brains of the operation,” Reliance combines application awareness, closed-loop orchestration and SLA automation to deliver optimal application performance, and Univa UD customers and prospects view it as a key part of their cloud or utility infrastructures.

Although no one is talking about grid computing, Univa UD’s Gordon Jackson says the company’s experiences with grid and large-scale distributing processing management feed directly into its success with Reliance, especially as it relates to resource management and distribution. Jackson is the technical director of the Data Center Business Division former virtualization evangelist at DataSynapse.

Brown agrees that a real cloud-like solution requires a significant understanding of grid concepts. “However, because people thought of grid as so specialized, it got a lot of baggage,” he explains. “So while a lot of the core technology is very relevant, a lot of the terminology and a lot of the old processes are not. In fact, they hinder the adoption of the technology for cloud.”

Ivan Casanova, vice president of product marketing at the aforementioned DataSynapse sees a connection, too, calling grid computing the starting point for cloud computing — “a proof point for shared and dynamic infrastructure.” A big part of cloud computing is the ability to scale based on demand, and grid computing middleware is a great method for doing so, he says. (Casanova also notes that SOA is the architectural model for cloud computing, and DataSynapse has customers deploying SOAs and using the company’s GridServer product to scale those services.)

On the data grid front, Oracle’s Cameron Purdy, vice president of Fusion Middleware, says, “Data grid technology … is almost essential in any transactional processing or other data-intensive system that would be deployed into a cloud environment. I can’t imagine how you would run a data-intensive application across any number of servers in that type of environment without the ability to share and coordinate access to and operate and react to changes and events occurring to that information.”

According to Platform Computing Chairman and CEO Songnian Zhou, his customers definitely see the grid-cloud connection as they move from HPC-focused enterprise grids to general-purpose, often virtualized, shared-services platforms. “They may not call it cloud, they may not call it on-demand datacenter, but they clearly are doing it,” he says.

The transition from grid to cloud, at least internally, Zhou says, is really a matter of evolution: the architectures and goals are the same, but the scope is different. As users move from HPC workloads to more generic workloads, they add components like J2EE middleware and hypervisors to enable more dynamic applications and increase mobility. “The tools and containers need to be brought to the plate, but [in terms of] fundamental architecture and approach, I don’t see much of a difference between grid and the cloud or on-demand or dynamic datacenter,” he says. “It’s a continuous evolution and expansion … away from the siloed client servers.”

Univa UD’s Jackson also sees this move happening. Even in HPC, he believes, grid is ripe to become a platform for serving multiple applications and classes of users. “[A]s soon as you start applying the intelligence to differentiate between the platinum customer and the bronze customer, or the applications … and services I need to execute on behalf of those customers, then I think you’ve taken your grid and you’ve turned it into a utility infrastructure,” he says. Moving discrete grids into one big, intelligent resource pool means the most bang for the buck for the corporation, he added.

Paul Strong, distinguished engineer at eBay Research Labs and active Open Grid Forum participant, isn’t even sure we should draw the distinction between so-called internal clouds and service-oriented grids. In either case, users are leveraging shared infrastructures and virtualization to achieve high utilization and application flexibility based on changes in workloads and business needs. Essentially, Strong says, users are solving the same problems with the same core technologies.

This is true even for eBay’s infrastructure, which Strong has explained as a grid for years. There are many “cloudy” aspects to eBay’s infrastructure, he says, including a heavily virtualized database architecture to allow for massive scalability, and global service delivery backed up by SLAs, continuous availability and security. “By some definition of the word,” he acknowledges, “I would say we’ve been doing aspects of cloud computing for a while.”

Second, Contrast

There are, however, differences between the grid and the cloud, especially, but not exclusively, where external cloud services are involved. Univa UD’s Brown makes a marked distinction between the two paradigms. For him, grid computing (on the enterprise front, at least) takes place inside the firewall. Apparently, Univa UD also notices grid computing’s HPC connotation, as the company has moved its Grid MP middleware to its HPC division, leaving (as noted above) the datacenter division to focus on Reliance and its automation capabilities.

A similar mindset seems to be present at Platform. Although not formally in place, Zhou says the company is moving toward two distinct foci — HPC and the datacenter. Regardless of the technological similarities, customers see grid computing being part of the former and cloud computing (to whatever degree the term arises) being part of the latter. As the datacenter division takes shape, Zhou says, “We will not emphasize grid much because for datacenters, I think, grid is foreign. It has too much connotation, it is tainted from the HPC or research and government space, and it’s too complex.”

At DataSynapse, the difference is very application-focused. Casanova believes grid addresses a specific class of applications, which have been successfully optimized using the company’s GridServer product. When customers wanted to run a more general class of applications on a shared, dynamic infrastructure, the result was DataSynapse’s FabricServer product. Going forward, both are part of the company’s greater application-focused cloud-like strategy.

“I don’t think people started out envisioning grid computing to be this seamless model where all these enterprise services or different types of applications were running in a cloud, they were universally accessible, they were technology-agnostic,” says Casanova. “I think they started that conversation around ‘I really need to scale up this application from a performance perspective, and I want to leverage commodity hardware and systems I already own to help me [experience] an order of magnitude improvement [in how] this important application executes.'” The cloud movement inside organizations, he adds, really has been driven by what they see Internet companies like Amazon doing with their infrastructures.

Further illustrating the fundamental differences between the paradigms, Casanova cites capabilities that must be added to grid solutions if they are to be repackaged and sold as cloud solutions: automated provisioning, horizontal scalability and visibility into utilization. For corporate users, he says, utilization insight helps them define policies to further automate and maximize resource usage. For cloud service providers, he says, usage data becomes the foundation for their chargeback models.

eBay’s Strong views grids and external clouds “less in terms of technology and more in terms of the way we think differently about business.” With cloud computing, he says, all a company needs to do is codify its one differentiating idea. The company can obtain everything else — infrastructure, non-critical business services, etc. — from the cloud.

Grid computing, Strong explains, was about building and optimizing infrastructure to run certain types of workloads. Cloud computing complements and advances that notion, but also helps companies move away from the “nitty-gritty” aspects, he says. It’s more about flexibly delivering services within an organization (internal cloud) or flexibly receiving commodity services (external cloud). “I think clouds take us to that next conceptual step of moving beyond … the weeds, where a lot of the grid work was, to [asking] ‘How do we move this closer to the business and deliver the right value for the business?'” Strong says. “The set of underlying technologies are, in essence, the same, but the discussion has changed.”


Up next: The second half of our look at the nexus of grid computing and cloud computing will focus on the future dynamic datacenter-oriented strategies of these old-guard grid computing vendors. We’ll look at how, if at all, they plan to leverage cloud computing hype as a marketing term, and how the idea of cloud computing actually relates to what they’re doing.

Subscribe to HPCwire's Weekly Update!

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

UCSD, AIST Forge Tighter Alliance with AI-Focused MOU

January 18, 2018

The rich history of collaboration between UC San Diego and AIST in Japan is getting richer. The organizations entered into a five-year memorandum of understanding on January 10. The MOU represents the continuation of a 1 Read more…

By Tiffany Trader

New Blueprint for Converging HPC, Big Data

January 18, 2018

After five annual workshops on Big Data and Extreme-Scale Computing (BDEC), a group of international HPC heavyweights including Jack Dongarra (University of Tennessee), Satoshi Matsuoka (Tokyo Institute of Technology), Read more…

By John Russell

Researchers Measure Impact of ‘Meltdown’ and ‘Spectre’ Patches on HPC Workloads

January 17, 2018

Computer scientists from the Center for Computational Research, State University of New York (SUNY), University at Buffalo have examined the effect of Meltdown and Spectre security updates on the performance of popular H Read more…

By Tiffany Trader

HPE Extreme Performance Solutions

HPE and NREL Take Steps to Create a Sustainable, Energy-Efficient Data Center with an H2 Fuel Cell

As enterprises attempt to manage rising volumes of data, unplanned data center outages are becoming more common and more expensive. As the cost of downtime rises, enterprises lose out on productivity and valuable competitive advantage without access to their critical data. Read more…

Fostering Lustre Advancement Through Development and Contributions

January 17, 2018

Six months after organizational changes at Intel's High Performance Data (HPDD) division, most in the Lustre community have shed any initial apprehension around the potential changes that could affect or disrupt Lustre Read more…

By Carlos Aoki Thomaz

UCSD, AIST Forge Tighter Alliance with AI-Focused MOU

January 18, 2018

The rich history of collaboration between UC San Diego and AIST in Japan is getting richer. The organizations entered into a five-year memorandum of understandi Read more…

By Tiffany Trader

New Blueprint for Converging HPC, Big Data

January 18, 2018

After five annual workshops on Big Data and Extreme-Scale Computing (BDEC), a group of international HPC heavyweights including Jack Dongarra (University of Te Read more…

By John Russell

Researchers Measure Impact of ‘Meltdown’ and ‘Spectre’ Patches on HPC Workloads

January 17, 2018

Computer scientists from the Center for Computational Research, State University of New York (SUNY), University at Buffalo have examined the effect of Meltdown Read more…

By Tiffany Trader

Fostering Lustre Advancement Through Development and Contributions

January 17, 2018

Six months after organizational changes at Intel's High Performance Data (HPDD) division, most in the Lustre community have shed any initial apprehension aroun Read more…

By Carlos Aoki Thomaz

When the Chips Are Down

January 11, 2018

In the last article, "The High Stakes Semiconductor Game that Drives HPC Diversity," I alluded to the challenges facing the semiconductor industry and how that may impact the evolution of HPC systems over the next few years. I thought I’d lift the covers a little and look at some of the commercial challenges that impact the component technology we use in HPC. Read more…

By Dairsie Latimer

How Meltdown and Spectre Patches Will Affect HPC Workloads

January 10, 2018

There have been claims that the fixes for the Meltdown and Spectre security vulnerabilities, named the KPTI (aka KAISER) patches, are going to affect applicatio Read more…

By Rosemary Francis

Momentum Builds for US Exascale

January 9, 2018

2018 looks to be a great year for the U.S. exascale program. The last several months of 2017 revealed a number of important developments that help put the U.S. Read more…

By Alex R. Larzelere

ANL’s Rick Stevens on CANDLE, ARM, Quantum, and More

January 8, 2018

Late last year HPCwire caught up with Rick Stevens, associate laboratory director for computing, environment and life Sciences at Argonne National Laboratory, f Read more…

By John Russell

Inventor Claims to Have Solved Floating Point Error Problem

January 17, 2018

"The decades-old floating point error problem has been solved," proclaims a press release from inventor Alan Jorgensen. The computer scientist has filed for and Read more…

By Tiffany Trader

US Coalesces Plans for First Exascale Supercomputer: Aurora in 2021

September 27, 2017

At the Advanced Scientific Computing Advisory Committee (ASCAC) meeting, in Arlington, Va., yesterday (Sept. 26), it was revealed that the "Aurora" supercompute Read more…

By Tiffany Trader

Japan Unveils Quantum Neural Network

November 22, 2017

The U.S. and China are leading the race toward productive quantum computing, but it's early enough that ultimate leadership is still something of an open questi Read more…

By Tiffany Trader

AMD Showcases Growing Portfolio of EPYC and Radeon-based Systems at SC17

November 13, 2017

AMD’s charge back into HPC and the datacenter is on full display at SC17. Having launched the EPYC processor line in June along with its MI25 GPU the focus he Read more…

By John Russell

Nvidia Responds to Google TPU Benchmarking

April 10, 2017

Nvidia highlights strengths of its newest GPU silicon in response to Google's report on the performance and energy advantages of its custom tensor processor. Read more…

By Tiffany Trader

IBM Begins Power9 Rollout with Backing from DOE, Google

December 6, 2017

After over a year of buildup, IBM is unveiling its first Power9 system based on the same architecture as the Department of Energy CORAL supercomputers, Summit a Read more…

By Tiffany Trader

Fast Forward: Five HPC Predictions for 2018

December 21, 2017

What’s on your list of high (and low) lights for 2017? Volta 100’s arrival on the heels of the P100? Appearance, albeit late in the year, of IBM’s Power9? Read more…

By John Russell

GlobalFoundries Puts Wind in AMD’s Sails with 12nm FinFET

September 24, 2017

From its annual tech conference last week (Sept. 20), where GlobalFoundries welcomed more than 600 semiconductor professionals (reaching the Santa Clara venue Read more…

By Tiffany Trader

Leading Solution Providers

Chip Flaws ‘Meltdown’ and ‘Spectre’ Loom Large

January 4, 2018

The HPC and wider tech community have been abuzz this week over the discovery of critical design flaws that impact virtually all contemporary microprocessors. T Read more…

By Tiffany Trader

Perspective: What Really Happened at SC17?

November 22, 2017

SC is over. Now comes the myriad of follow-ups. Inboxes are filled with templated emails from vendors and other exhibitors hoping to win a place in the post-SC thinking of booth visitors. Attendees of tutorials, workshops and other technical sessions will be inundated with requests for feedback. Read more…

By Andrew Jones

Tensors Come of Age: Why the AI Revolution Will Help HPC

November 13, 2017

Thirty years ago, parallel computing was coming of age. A bitter battle began between stalwart vector computing supporters and advocates of various approaches to parallel computing. IBM skeptic Alan Karp, reacting to announcements of nCUBE’s 1024-microprocessor system and Thinking Machines’ 65,536-element array, made a public $100 wager that no one could get a parallel speedup of over 200 on real HPC workloads. Read more…

By John Gustafson & Lenore Mullin

Delays, Smoke, Records & Markets – A Candid Conversation with Cray CEO Peter Ungaro

October 5, 2017

Earlier this month, Tom Tabor, publisher of HPCwire and I had a very personal conversation with Cray CEO Peter Ungaro. Cray has been on something of a Cinderell Read more…

By Tiffany Trader & Tom Tabor

Flipping the Flops and Reading the Top500 Tea Leaves

November 13, 2017

The 50th edition of the Top500 list, the biannual publication of the world’s fastest supercomputers based on public Linpack benchmarking results, was released Read more…

By Tiffany Trader

GlobalFoundries, Ayar Labs Team Up to Commercialize Optical I/O

December 4, 2017

GlobalFoundries (GF) and Ayar Labs, a startup focused on using light, instead of electricity, to transfer data between chips, today announced they've entered in Read more…

By Tiffany Trader

How Meltdown and Spectre Patches Will Affect HPC Workloads

January 10, 2018

There have been claims that the fixes for the Meltdown and Spectre security vulnerabilities, named the KPTI (aka KAISER) patches, are going to affect applicatio Read more…

By Rosemary Francis

HPC Chips – A Veritable Smorgasbord?

October 10, 2017

For the first time since AMD's ill-fated launch of Bulldozer the answer to the question, 'Which CPU will be in my next HPC system?' doesn't have to be 'Whichever variety of Intel Xeon E5 they are selling when we procure'. Read more…

By Dairsie Latimer

  • arrow
  • Click Here for More Headlines
  • arrow
Share This