Cray
HPCwire

Since 1986 - Covering the Fastest Computers
in the World and the People Who Run Them

Language Flags

Visit additional Tabor Communication Publications

Datanami
Digital Manufacturing Report
HPC in the Cloud
Green Computing Report

Tabor Communications
Corporate Video

Grid Software Creates Virtual Windows Supercomputer


What if you could combine the benefits of virtualization, grid and cloud computing to accelerate Windows-based applications? An Israeli company, Xoreax, is doing just that. We spoke with Xoreax at SC12 in Salt Lake City, Utah, last week to learn more about their offering.

Xoreax got its start back in 2002 and for the last 10 years, they've been accelerating software in the Windows environment, using distributed, aka grid, computing technology. Their IncrediBuild-XGE (Xoreax Grid Engine) software uses a unique technology called process level virtualization to create a virtual HPC machine. The software harnesses unutilized CPUs to create a private grid. It runs on a company's existing Windows infrastructure and extends into the public cloud if more resources are required.

Typical server and CPU utilization rates are notoriously low; IncrediBuild finds all that spare compute capability and exposes it.

"It's like having virtual HPC – every workstation in this network can scale out to tap into the amount of resources that is in the entire network," says CTO Dori Exterman.

"We have an agent in every machine and create a private grid out of all these machines and can also scale out to the cloud. The solution is currently Windows-only, and there aren't many Windows solutions for this kind of cycle scavenging."

Unlike other grid and virtualization solutions, IncrediBuild works out of the box. Traditional grid solutions require the user to change the source code and target an API. This extra work makes sense for very demanding applications like real-time financial trading, but Xoreax is targeting more general-purpose workloads that are also process-intensive.

The solution combines grid and virtualization, but instead of virtualizing entire machines like VMware does, the technology virtualizes a single process.

"We can take a process and virtualize it on the network. When the process needs some files from the local machine, the software acts like a middleman between the process and the operating system and we can take all that environment that the process requires on demand from the initial machine, cache it on the remote machine and give the process everything it needs in order to work on the remote machine," explains Exterman.

"The process does not care where it's running because we will create all the environment for it. You don't need to install any software on the remote machines," he adds.

Traditional HPC usually requires costly investments of time and resources in hardware, training and expertise, whereas process virtualization leverages the customers own resources. Potential customers are interested in ease of implementation and integration and they don't want to be vendor-bound, the CTO notes, adding that integration takes less than an hour and requires only a small 2-line xml file.

Perhaps owing to Xoreax's niche as a code-build accelerator, the company is not that widely known despite healthy deployment numbers. Over 100,000 users from some 50 countries rely on XGE-based products, and customers include big-name corporations like Cisco, NVIDIA, Google, Microsoft, eBay and IBM.

Under the OEM model, a software vendor integrates with IncrediBuild-XGE and then provides the product to their end users. One such example is the large diamond grading outfit Sarin, which uses the software to accelerate diamond analysis for tens of thousands of customers. Instead of running all the simulations on a single dedicated server they can scale out across all their in-house resources and into the cloud. With the XGE client, Sarin has achieve a 20x speedup in system performance. The time required for complex analyses has shrunk from 40-60 hours down to 2-4 hours, while simpler 1-2 hour analyses now complete in 10-15 minutes.

As with most grid or cloud solutions, IncrediBuild is intended for applications that are compute-intensive as opposed to I/O-intensive. It's well-suited for computational tasks that take a lot of time, such as batch processing or near real-time. Gaming, financial services and manufacturing are some key verticals.

While early use cases revolved around computation acceleration, the software development space has evolved into their main customer base. IncrediBuild reduces the time of many development environment tasks, including code analysis, unit testing, QA script, stress tests and code verification. The company states it is the most popular off-the-shelf accelerator for Visual Studio code builds.

Another interesting IncrediBuild-XGE use case is the follow-the-moon scenario, where two manufacturing centers located in opposite hemispheres distribute their resources to balance workloads.

On the OEM side, IncrediBuild-XGE has added support for GPUs. This capability offers a lot of potential to engineers working with graphics-intensive workloads, like rendering or CFD. Running an application on a remote server with a dedicated discrete GPU can take minutes instead of hours versus running on a local workstation with an on-board GPU.

IncrediBuild-XGE is compatible with virtually any IaaS offering. Because the engine works with VMware, it runs everywhere you can mount a machine and have VMware run it as part of your subnet. During times of high demand, users can add virtually unlimited resources to their IncrediBuild grid.

On the subject of competition, both the CEO Eyal Maor and the CTO respond that the process virtualization technology is unique and that there aren't really any comparable offerings. When I inquire about Windows-based grid vendor Digipede specifically, they acknowledge the similarities, but state that they "don't run into them."

As for their future roadmap, Xoreax doesn't want to reveal too much, but the company hints at several announcements that are coming down the pike that could significantly expand their user ecosystem, including plans around Linux and tablets.

Xoreax offers separate pricing terms for its OEM and off-the-shelf products. The cost of the OEM version is negotiated on a per-customer basis, while the off-the-shelf pricing is per agent and based on the total number of cores that the agent is using. The company also offers different product packages, depending on the type of acceleration desired. Interested users can download a trial license, which gives them a chance to kick the tires and decide if the solution is right for them.

June 18, 2013

June 17, 2013

June 14, 2013

June 13, 2013

June 12, 2013

June 11, 2013

June 10, 2013

June 07, 2013

June 06, 2013

June 05, 2013


Most Read Features

Most Read Around the Web

Most Read This Just In


Short Takes

Supercomputers: Not Always the Best for Big Data

Jun 18, 2013 | The world's largest supercomputers, like Tianhe-2, are great at traditional, compute-intensive HPC workloads, such as simulating atomic decay or modeling tornados. But data-intensive applications--such as mining big data sets for connections--is a different sort of workload, and runs best on a different sort of computer.
Read more...

Gordon Flashes Its Versatility in HPC Workloads

Jun 18, 2013 | Researchers are finding innovative uses for Gordon, the 285 teraflop supercomputer housed at the San Diego Supercomputer Center (SDSC) that has a unique Flash-based storage system. Since going online, researchers have put the incredibly fast I/O to use on a wide variety of workloads, ranging from chemistry to political science.
Read more...

Supercomputers: Still the King of the HPC Hill

Jun 17, 2013 | The advent of low-power mobile processors and cloud delivery models is changing the economics of computing. But just as an economy car is good at different things than a full size truck, an HPC workload still has certain computing demands that neither the fastest smartphone nor the most elastic cloud cluster can fulfill.
Read more...

TACC Longhorn Takes On Natural Language Processing

Jun 14, 2013 | For all the progress we've made in IT over the last 50 years, there's one area of life that has steadfastly eluded the grasp of computers: understanding human language. Now, researchers at the Texas Advanced Computing Center (TACC) are utilizing a Hadoop cluster on its Longhorn supercomputer to move the state of the art of language processing a little bit further.
Read more...

Titan Didn't Redo LINPACK for June Top 500 List

Jun 13, 2013 | Titan, the Cray XK7 at the Oak Ridge National Lab that debuted last fall as the fastest supercomputer in the world with 17.59 petaflops of sustained computing power, will rely on its previous LINPACK test for the upcoming edition of the Top 500 list.
Read more...

Sponsored Whitepapers

Best Practices in Big Data Storage

05/10/2013 | Cleversafe, Cray, DDN, NetApp, & Panasas | From Wall Street to Hollywood, drug discovery to homeland security, companies and organizations of all sizes and stripes are coming face to face with the challenges – and opportunities – afforded by Big Data. Before anyone can utilize these extraordinary data repositories, however, they must first harness and manage their data stores, and do so utilizing technologies that underscore affordability, security, and scalability.

Progress in Parallel: the Bull Parallel Programming Center

04/15/2013 | Bull | “50% of HPC users say their largest jobs scale to 120 cores or less.” How about yours? Are your codes ready to take advantage of today’s and tomorrow’s ultra-parallel HPC systems? Download this White Paper by Analysts Intersect360 Research to see what Bull and Intel’s Center for Excellence in Parallel Programming can do for your codes.

Sponsored Multimedia

HPCwire Live! Atlanta's Big Data Kick Off Week Meets HPC

Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?

Webinar: Mellanox Virtual Modular Switch, the Most Efficient 40GbE Aggregation Switch Solution

Join our webinar to learn how IT managers can migrate to a more resilient, flexible and scalable solution that grows with the data center. Mellanox VMS is future-proof, efficient and brings significant CAPEX and OPEX savings. The VMS is available today.

Atlanta's Big Data Kick Off Week Meets HPC Cray Exxact

HPC Job Bank


Featured Events






  • November 17, 2013 - November 22, 2013
    SC'13
    Denver, CO
    United States


HPCwire Events