ARM Yourself

ARM servers from Pogo Linux

With ARM processors becoming so prevalent in our cell phones, it’s easy to forget they can excel in HPC too. The Acorn RISC Machine (later known as “Advanced RISC Machine” or ARM) architecture’s simplified design results in high density, exceptional power efficiency, and minimized heat. While these characteristics are ideal for low-power mobile devices, they can also be used to great advantage in the data center.

An Affordable Alternative to x86

Case in point: The Pogo Linux ARM R150-T62, pre-configured with dual 48-core 64-bit ARM processors, up to 1TB of DDR4 memory, and multiple 40GbE and 10GbE network interfaces. The high core counts and extended memory bandwidth make this system-on-chip (SoC) server the best in class for performance/$ and performance/watt. The ARM R150 is truly an affordable alternative to x86-based systems.


ThunderX_Processors• Public & Private Cloud
• Content Delivery
• Web Caching
• Web Search
• Social Media Data Analytics

Cavium ThunderX_CP™ Processors

The ThunderX_CP™ product family is the best in class 64-bit ARMv8 Data Center & Cloud Processors, offering an unprecedented level of integration and industry leading SoC performance. These processors are based on highly efficient full custom processor cores
designed by Cavium in 28nm process technology under architectural license from ARM. They are fully compliant with ARMv8 architecture as well as ARM’s Server Base System Architecture (SBSA) standard.

 cavium thunderx diagram

Affordable Processing Power to Spare

What could you accomplish with 96 cores of inexpensive processing power running in a cool and efficient rackmount server? Contact the Pogo Linux Sales Engineering Team to learn more about this exciting new platform. They can assess the needs of your organization and put together the perfect ARM-based solution for you.

Find your ideal ARM server

To learn about how Pogo can help you tackle the technological challenges your organization is facing, please call us at 888-828-POGO, or email

Posted in Rackmount Servers, Server Solutions | Tagged , | Leave a comment

Intel introduces new Kaby Lake Core i7 CPUs

7th Gen Intel Core i7 Now Available

All compatible Pogo Linux workstations can now be configured with Intel’s brand new 7th gen Core i7 processors (aka Kaby Lake). Iterating on their outstanding Skylake architecture, Intel has made a number of optimizations. Here is a quick rundown of features for these new processors.

●  Improved 14 nm+ manufacturing process

●  Faster clock speeds – up to 4.2 GHz

●  Higher turbo frequencies – up to 4.5 GHz

●  Quicker clock speed changes – provides enhanced power savings

●  Improved graphics core – allows for faster video decoding

●  Support for Intel Optane technology – Intel’s new breakthrough memory architecture

Contact us today and talk to one of our sales engineers about how you can benefit from this new and improved CPU architecture.

7th Gen Intel Core i7 Workstations

To learn about how Pogo can help you tackle the technological challenges your organization is facing, please call us at 888-828-POGO, or email

Posted in General, Press Releases, Workstations | Tagged , , , | Leave a comment

Global Namespaces, Simplified

Pogo Linux Storage Director Q featuring QuantaStor by OSNEXUS

As storage environments grow and appliances are added to expand capacity, it becomes increasingly difficult for users to manage their data across disparate sets of storage appliances. This, in turn, slows productivity. One traditional approach to solving this data management problem has been to use scale-out NAS solutions that provide visibility to all NAS shares from all appliances that are part of a given scale out volume.

Scale-Out NAS Limitations

Scale-out NAS solutions, however, introduce a number of problems including spindle-contention, large failure domains, poor performance for small block IO, higher latency and higher up-front costs due to the need for the additional high-speed 40GbE+ backend network to interconnect appliances.

So how does a storage administrator keep the performance benefits of a highly-available scale-up NAS system while also getting the ease of management that comes with the global-namespace or single namespace capability of a scale-out NAS system? One answer is to combine control-plane and data-plane storage management capabilities across a single global namespace. This namespace groups shares together into a globally accessible folder across selected appliance groups.

QuantaStor by OSNexus: Combining the Data- and Control-Plane for Easier Management

One Software Defined Storage solution that can provide this feature set is QuantaStor by OSNexus. QuantaStor combines the best of these two technologies through innovations in both the SDS control-plane and data-plane (see figure below). QuantaStor’s control-plane solves the ease-of-management issues through its unique grid management technology enabling up to 64 appliances to link together across sites and datacenters providing homogeneous management of all appliances as a single grid. QuantaStor’s data-plane supports global namespaces, also known as Network Share Namespaces, which implement Microsoft DFS and NFSv4 referral technologies to redirect SMB/CIFS clients to network share locations and can work together with Highly-Available (HA) storage pools. This lets users group together network shares across any number of appliances in a QuantaStor grid and present them from all appliance nodes as if they were scale-out NAS.

Storage Director Q NFS SMBCommon use cases for global namespaces include media production workflows that lets IT managers distribute files in a way that achieves the best performance and capacity utilization per system. Artists, editors and producers are able to access their digital assets via a single, logical namespace, even across geographies. The global namespace reduces human error, improves productivity and streamlines media and entertainment workflows enabling NTV networks, studios and VFX companies to accomplish more at a lower cost while eliminating complexity.

Pogo Linux Storage Director

The Pogo Linux Storage Director Q Series of systems, provide the perfect overall solution to achieving a successful global namespace deployment. By combining an optimal hardware configuration with the QuantaStor SDS, Pogo has a complete solution that solves the traditional issues of scale-out storage by providing both high performance and ease of management. This is all backed by Pogo’s award winning support and warranty coverage.

Want more information about how the Storage Director Q  series and QuantaStor can solve your data storage needs? Give our sales engineers a shout at 1-888-828-7646 or drop us an email at

Pogo Linux Storage Director

To learn about how Pogo can help you tackle the technological challenges your organization is facing, please call us at 888-828-POGO, or email

Posted in Press Releases, Storage, Storage Solutions | Tagged , , , , | Comments Off on Global Namespaces, Simplified

VR is for Linux, too! (Ready to try it?)

VR is now running on Linux

With the recent beta release of Steam VR for Linux, lots of developers and gamers are eager to take the plunge into VR apps in the Linux environment. Naturally, the Pogo Engineering Team got right to work configuring it on test systems here at Pogo HQ. Interested in giving this a try yourself? Read on…

Testing Steam VRWe’d be remiss if we didn’t mention a small caveat. As exciting as this is, we should warn you there is still a ways to go before Steam VR for Linux is ready for prime time. We recommend you dive into this only if you’re up to the challenge of dealing with some potential hiccups (and lots of reboots).

If you’re ready to try VR in Linux right away, here are the basic steps to get it up and running. These instructions come straight from our engineering team. They got Steam VR running on Ubuntu 16.04 on a Pogo Linux Tempest Mini VR Edition Workstation. Let’s get started!

1. Download the Steam VR Linux beta:
Steam VR

2. Remove the nouveau drivers and install Nvidia drivers. (Skip this if Ubuntu is freshly installed.)

sudo apt-get remove nvidia* && sudo apt autoremove

3. Install some packages for building the kernel:

sudo apt-get install dkms build-essential linux-headers-generic

4. Block and disable nouveau kernel driver:

sudo vi /etc/modprobe.d/blacklist.conf

5. Insert the following lines into the blacklist.conf:

blacklist nouveau
blacklist lbm-nouveau
options nouveau modeset=0
alias nouveau off
alias lbm-nouveau off

6. Save and exit.

7. Disable the Kernel nouveau by typing the following commands (nouveau-kms.conf may not exist – it is not a problem if it doesn’t):

sudo echo options nouveau modeset=0
sudo tee -a /etc/modprobe.d/nouveau-kms.conf

8. Build the new kernel with this command:

sudo update-initramfs -u

9. Reboot.

10. Install Nvidia driver 375.27.13.

a. Add the repository to get the correct version of the NVIDIA Developer Beta Driver:

sudo add-apt-repository ppa:mamarley/nvidia-dev


b. Open “additional drivers” then pick 375.27.13.

11. Install steam and opt into steam beta client (Left click view, then settings, then under account select “beta participation”).
steam client beta

12. Install the SteamVR tool and opt into the beta (Right click SteamVR, select properties, click the beta tab, and select the beta from the dropdown menu).

Steam VR Setup Complete

13. Test and enjoy! If you’d like to share your experience with us, or have any questions about VR development, please let us know.

Steam VR Powered Systems from Pogo Linux

To learn about how Pogo can help you tackle the technological challenges your organization is facing, please call us at 888-828-POGO, or email

Posted in VR Virtual Reality, Workstations | Tagged , , , , , | Comments Off on VR is for Linux, too! (Ready to try it?)

Pogo Linux and Scrypt: Rising to the challenges of HIPAA with encrypted SSDs

Pogo Customer Profile: Scrypt, Inc.

Since 1998, Scrypt, Inc. has been a leading provider of hosted services to regulated markets, specifically healthcare, and mortgage origination. The customer-centric company offers a vast product line of secure, reliable services, from cloud-based faxing and secure messaging to regulatory compliance and document management. Their customers demand strict compliance with HIPPA and financial regulations, so it is imperative that Scrypt deploy hardware and software that meets these stringent standards.

To implement the wide variety of services and applications they provide, Scrypt offers many different hardware configurations. This enables their customers to maximize performance and minimize cost. Scrypt stakes their business reputation on the hardware they use, so they need a reliable hardware provider who can also serve as a “trusted advisor,” providing guidance and recommendations for specific workload scenarios.

The Challenge: Finding a Rock Solid Supermicro Partner

From the beginning, Scrypt has depended on the reliability and simplicity of systems built on Supermicro server platforms. The cost effective price point of Supermicro hardware, combined with numerous component options, made it the logical choice. However, with all that flexibility, Scrypt needed the right integrator to guide them through all of the Supermicro server options. Pogo was the ideal partner to design and build Scrypt’s servers, as for the past 18 years they have delivered quality solutions built on Supermicro platforms. Scrypt CTO Nick Basil summed it up succinctly: “What drives us to Pogo is that they specialize in Supermicro-based solutions — they do it really, really well.”

Supermicro systems from Pogo LinuxAfter working with the Pogo sales and support staff on a few projects, it became clear to Basil that he had made the right choice in engaging with Pogo. When designing a solution, Scrypt will “…have a conversation with Pogo about what the application is, what the scenario is, and ask them to provide recommendations. Sometimes we’ll procure stuff for customers and ship it pre-deployed, already set up with our application, and other times we’ll have the customers go right to Pogo,” said Nick. This level of trust and expertise has proved to be an ideal combination for Scrypt. They are confident about sending their own customers directly to Pogo, knowing they will get high-quality hardware solutions, perfectly tailored to their needs.

HIPAA Compliance with Encrypted Solid State Drives

Scrypt especially appreciated Pogo’s value as a trusted advisor when Scrypt decided to Micron SSDmove away from spinning disks in favor of Solid State storage. With advantages like reduced power consumption and increased speed, Scrypt knew they needed to make a shift in their primary storage medium. They had purchased increasing numbers of SSD-based systems over the past few years, but the transition had been challenging. With the SSD market changing rapidly, it was difficult to consistently determine the best product for a given workload. In addition, Scrypt required data-at-rest encryption to make sure they met rigid HIPAA compliance standards. Basil relied on Pogo for guidance. “I can only see a certain number of variables, so I’ll call Pogo and say, ‘Here’s the load the SSD is going to be under, what do you recommend? Micron? Intel?’ We discuss the budget, and they advise me on the best choices for each scenario,” said Basil. He depends on Pogo to ensure that Scrypt always deploys the correct product for their services.

Attention to Detail

In addition to doling out expert advice, Scrypt’s confidence in Pogo also stems from the impeccable build quality and attention to detail evident in every system they sell. “We’re really particular about how systems are assembled — cables, fans, everything. Everything we get, we open up and verify. We’re going to take the lid off and make sure the cables are tight and tied. Pogo’s servers are tied, zipped, and laid out so that if anybody were to open it up they’d think, ‘This was assembled very well.’ So, that’s why we go with Pogo. I know when I get something, it’s been put together to the highest standards,” said Basil.

Moving Forward

As Scrypt builds out its infrastructure, Pogo continues to be an important part of their planning and purchasing process. Whether it is transitioning to new technologies or guidance on current ones, Pogo is always there with the correct hardware and expert advice to make the process easy. For Scrypt, having a reliable partner is invaluable. “The most important point about our relationship with Pogo Linux is that we don’t have to do things twice. When we bring in something from Pogo, it works. Pogo always takes care of us,” said Basil.

Supermicro Server from Pogo Linux

To learn about how Pogo can help you tackle the technological challenges your organization is facing, please call us at 888-828-POGO, or email

Posted in Case Studies, General, Server Solutions | Tagged , , | Comments Off on Pogo Linux and Scrypt: Rising to the challenges of HIPAA with encrypted SSDs

Now Available: The NVIDIA GeForce GTX 1080 Ti

You may recall our recent blog post introducing the NVIDIA TITAN X™ GPU, which made desktop supercomputing a much more affordable proposition. Well, NVIDIA has just raised the bar again with the GTX 1080 Ti.

The GTX 1080 Ti looks to be a compelling alternative to the Titan X. It’s much more affordable, and in many aspects, has superior performance. Equally importantly, NVIDIA has not imposed any performance limits on the 1080 Ti to prevent competition with their more expensive Titan X. As a result, the 1080 Ti should be very popular for GPU computing. Unless a user’s workload demands the additional 1GB of VRAM found in the Titan, the GTX 1080 Ti could be the smart choice for GPU computing.

The GTX 1080 Ti is now available. Within the next week or two, we will be doing extensive benchmarking of this new GPU in our Velocity D58PSC Workstation. Check back here for a future blog post where we will quantify the performance differences between the GTX 1080 Ti and Titan X.

Check out all of the Pogo Linux GPU Workstations.

Configure your Velocity D58PSC with the NVIDIA TITAN X Pascal

To learn about how Pogo can help you tackle the technological challenges your organization is facing, please call us at 888-828-POGO, or email

Posted in Rackmount Servers | Comments Off on Now Available: The NVIDIA GeForce GTX 1080 Ti

Parallel Performance – Intel Xeon Phi Systems

Massively Parallel Intel Xeon Phi Systems

Today’s high performance computing applications are becoming as resource-intensive and complex as they are diverse. Across disciplines such as Earth Sciences, Financial Services, Physics, and Machine Learning, demands for CPU power and memory bandwidth have increased exponentially. Within this set of workloads a common component is the need for highly parallel processing to maximize performance. However, up until now there hasn’t been a good solution to run these applications.

Intel Xeon Phi ProcessorOne solution was to to use regular Intel® Xeon™ processors. However, although extremely robust, the processor architecture is designed for general IT workloads. These processors support advanced features for virtualization and security that work great for central IT and storage activities. However, these special features are not required for highly parallel workloads. This means that a lot of the core processor functionality is wasted.

On the other end of the spectrum, there are highly specialized GPU processors specifically designed for parallel processing (CUDA). Unfortunately, these processors are not bootable and require legacy code to be rewritten for an entirely new architecture. This can dramatically increase development costs and the length of time it takes to get a model up and running.

With this in mind, Intel has created a new HPC processor that lowers the barrier to entry, without sacrificing performance. The Intel Xeon Phi™ processor is a bootable host processor delivering massive parallelism and vectorization in support of the most demanding high-performance computing applications. By using an x86 processor architecture, the Intel Xeon Phi let’s a developer run legacy applications originally written for the Intel Xeon platform with minimal changes. This dramatically cuts down on development costs and time to market — or time to insight.

Download the benchmark results

Optimized for Parallel Performance

The new Intel Xeon Phi bootable host processor offers an integrated architecture for powerful, highly parallel performance that will blaze the trail to greater innovation and impact for today’s most-demanding High Performance Computing applications. With up to 72 out-of-order cores, this processor can deliver over 3 teraFLOPS (floating-point operations per second) of double-precision calculations.

Reduced Power Consumption

The many core architecture of the new Xeon Phi processors enables significant improvements in power consumption. The higher degree of parallelism in these processors results in greater compute per unit of energy consumed for highly-parallel applications.

Simplified Code Optimization

In the past, parallel CPU architectures required significant code re-writes in order to reap maximum performance benefits. This processor is binary-compatible with Intel Xeon processors, which allows it to run any x86 workload, optimizing asset utilization across the data center.

Deploy Your Own Custom Intel Xeon Phi System

Pogo Linux is introducing a new line of systems featuring the Intel Xeon Phi processor. Whatever your workload, our engineers can put together the ideal system for your needs. Now you’ve got a way to make it all happen.

Configure your Xeon Phi based system from Pogo Linux

To learn about how Pogo can help you tackle the technological challenges your organization is facing, please call us at 888-828-POGO, or email

Posted in Rackmount Servers, Server Solutions, Workstations | Comments Off on Parallel Performance – Intel Xeon Phi Systems

Dense and smart at the same time?

Density and intelligence need not be mutually exclusive. For instance, take the Pogo Linux J4600S storage enclosure. With this solution, an IT admin can decrease both their storage footprint and their power consumption — saving precious power and rack space in the data center. Using a combination of an advanced JBOD design and HGST Ultrastar® drives, the J4600S provides a superior JBOD solution.

High Capacity • Low Power Consumption

When data centers need storage expansion that delivers low power and high capacity with high-availability, the J4600S storage enclosure delivers. Based on the HGST 4U60 Storage Enclosure, this high-density, scalable, and cost-effective design includes up to 60 HGST Ultrastar® 3.5-inch  SAS hard drives modules in a 4U enclosure. Utilizing helium to fill the platter chamber instead of air, HGST Ultrastar hard drives are 47% more efficient (WATTS/TB) than conventional units. This energy efficiency combined with 12Gb/s SAS performance, high availability (HA), hot-swappable components, and a 5 year bumper-to-bumper warranty delivers the perfect storage solution for a variety of data center projects.

One JBOD, Many Uses

The J4600S has proven to be an extremely versatile unit for a wide range of use cases.  With the option to use up to 8 SSD devices per unit, there are numerous caching and tiering configurations available. This JBOD has been extensively tested with our two most common Software Defined Storage solutions (Nexenta and QuantaStor) and has proven to be a perfect solution for both regular and HA environments. In addition, we have created many ‘recipes’ in our lab for such applications as Ceph, Swift, and virtualized environments. The J4600S has proven to be so versatile and reliable, it is our go to storage enclosure for all our new enterprise class storage projects.

Dense and Smart

When it comes to storage, it is important to get the best solution for the job. That’s why we love the J4600S. Not only is it one of the densest solutions available on the market, it has the smarts to provide low power consumption and the enterprise features that today’s data centers demand. Take a look here, or give us a call or email so we can help configure your storage project today.


Download the datasheet (pdf)

•  Up to 60 integrated HGST Ultrastar® drive modules

•  Available with 512e or 4Kn sector size (by part number)
•  Available option to configure up to 8 Ultrastar SSDs to create separate performance tiers within the enclosure
•  Available with SAS drives for HA applications or with SATA drives for cost-optimized applications

•  Host connectivity using 2 x SAS QSFP+ receptacles per IO Module; optional connectivity cables are available

•  QSFP+ to HD mini SAS cables for host
•  QSFP+ to QSFP+ cables for daisy-chaining up to 4 enclosures

•  Hot Swappable: two power supplies (with integrated fans), drive modules, and IO modules
•  Fully compliant with SAS 3.0 specification for operation up to 12Gbps
•  Optional cable management arm eliminates tangled cables during servicing of hot-swappable CRUs
•  SCSI Enclosure Services (SES-3)
•  Comprehensive 5-year warranty on the entire assembly (not just the drives)

Configure your J4600S now

To learn about how Pogo can help you tackle the technological challenges your organization is facing, please call us at 888-828-POGO, or email

Posted in Press Releases, Rackmount Servers, Storage, Storage Solutions | Comments Off on Dense and smart at the same time?

NVMe – moving beyond the dark ages of storage

Embrace the non-volatile future

For maximum IOPS and minimum latency without a ridiculous power bill, it’s time for a better host controller interface to your flash storage! Thanks to the NVMe™ (Non-Volatile Memory express) specification, flash devices can now use the speed and bandwidth of a PCIe connection, without incurring the overhead of emulating a SAS interface. The I/O performance increase is up to 8x the speed of traditional SAS3 with 6x the power savings! The benefits to a desktop system are substantial, so imagine the impact at the enterprise level. NVMe storage pays incredible dividends when scaled up for deployments such as cloud hosting, relational databases, and online transaction processing.

About NVM Express

NVMeAs flash devices have become faster and more dense, they have also been held back by a legacy SAS interface. This interface was originally optimized for spinning disk which has much different characteristics than flash memory. Because of this, the SAS interface has always been a bottleneck for flash storage. NVM Express (NVMe) or Non-Volatile Memory Host Controller Interface Specification (NVMHCI) was specifically created as an optimal interface for the properties of flash storage. For example NVMe can handle 65,000 queues of data, each with 65,000 commands, compared to SAS that has a single queue with a capacity of 32 commands. In addition NVMe supports a streamlined set of only seven commands — far less than the SAS interface. Bottom line, this makes NVMe a far more efficient interface for flash storage, providing lower latency for the most demanding use cases.

Iris NV2248

Iris NV2248For the maximum savings of space, power, and cooling, look no further than the incredibly dense Iris NV2248. The patented Riser Bay design allows for double the number of 2.5″ drives in a 2U form factor. The 48 NVMe SSDs use 32 PCI-E 3.0 lanes from two Intel Xeon™ processors. The other 8 lanes are enabled by an innovative matrix of 4 PCI-E switches. Take a closer look at the Pogo Linux Iris NV2248.

• 48x 2.5″ Hot-Swap NVMe Drive Bays
• Dual Intel Xeon E5-2600 v4 Processors
• 1600W Redundant Power Supplies
• 2U Rackmount Form Factor

Iris NV2224

Iris NV2224The Iris NV2224 represents an ideal balance of throughput and capacity in a 2U chassis. It will host up to 24x hot-swap NVMe SSDs, all accessible from the front panel. Configure your own Iris NV2224.

• 24x 2.5″ Hot-Swap NVMe Drive Bays
• Dual Intel Xeon E5-2600 v4 Processors
• 1600W Redundant Power Supplies
• 2U Rackmount Form Factor

Iris NV1210

Iris NV1210Rounding out the collection of NVMe servers with maximum flexibility is the Iris NV1210. This 1U rack mount server sports a dense front panel with 10x 2.2″ hot-swap NVMe drive bays. Redundant 1000W power supplies ensure uninterrupted world-class performance. See for your yourself – outfit your own Iris NV1210.

• 10x 2.5″ Hot-Swap NVMe Drive Bays
• Dual Intel Xeon E5-2600 v4 Processors
• 1000W Redundant Power Supplies
• Compact 1U Rackmount Form Factor

Configure your Pogo Linux NVMe Server

To learn about how Pogo can help you tackle the technological challenges your organization is facing, please call us at 888-828-POGO, or email

Posted in Rackmount Servers, Server Solutions, Storage, Storage Solutions | Tagged , , , , , , , | Comments Off on NVMe – moving beyond the dark ages of storage

Is the NVIDIA Titan X up to the Challenges of Deep Learning?

NVIDIA Titan X with Pascal

NVIDIA has recently introduced the new TITAN X™, the biggest and most advanced GPU ever built. Blaise Pascal said, “Imagination decides everything.” This certainly applies to the hardware engineering team at NVIDIA. Not content with the already “irresponsible amount of performance” of their GTX 1080, they imagined breaking the 10 teraflops barrier for computing on a single chip. The decidedly impressive result is the new NVIDIA TITAN X with Pascal™.

NVIDIA TITAN X with Pascal

NVIDIA Tesla ArchitectureWhen tackling deep learning challenges, nothing short of ultimate performance will get the job done.  The latest Titan X is the first GPU to consider. The Titan X with Pascal is built atop NVIDIA’s new upsized GP102 architecture. Highlights of the GP102 include:

• 30 Streaming Multiprocessors
• 11 TFLOPS FP32
• 3584 CUDA Cores at 1.53GHz
• 12 GB GDDR5x memory (480 GB/s)
• 12,000,000,000 Transistors
• NVLink High-Speed SLI Interconnect

The Velocity D58PSC – a Desktop Supercomputer

Velocity D58PSC with Titan X CutawayOne might expect that computational learning projects would require racks and racks of dense compute nodes with lightning fast interconnects. However, technologies like the new NVIDIA Titan X bring High Performance Computing (HPC) right to your desktop. Pogo provides many researchers with their own Personal Supercomputers, such as the Pogo Linux Velocity D58PSC Workstation.

We recently built one of these high end workstations for an uncompromising client researching machine learning. Pushing the boundaries in a deep learning project, this researcher is using TensorFlow, one of the products from the Google Brain Team within the Google Machine Intelligence organization. One aspect of TensorFlow is that it works best with GPU computing. However, even with GPUs, performance increases are sublinear. As a result, achieving top performance requires top-of-the-line GPUs. This is the ideal use case for the new Titan X.

Our client required no less than four Titan X cards installed in their new system. Fortunately, the Velocity D58PSC provided the ideal platform for this application.

With Dual Intel Xeon E5-2600 v4 series processors and up to 1TB of DDR4 memory, the D58PSC will extract maximum performance from these new GPUs. Topping off the configuration with up to 8x SAS or SATA SSDs will ensure there will be no I/O bottlenecks throughout the system. In cutting edge research, having the right equipment is key. By combining the raw CPU and I/O power of the D58PSC with Titan X GPUs, this researcher has a recipe for success.

To find out more about GPU Computing and what it can do for your projects, Contact the Pogo Sales Engineering team team to learn more about these extraordinary technologies.

Configure your Velocity D58PSC with the NVIDIA TITAN X Pascal

To learn about how Pogo can help you tackle the technological challenges your organization is facing, please call us at 888-828-POGO, or email

Posted in Rackmount Servers, Workstations | Tagged , , | Comments Off on Is the NVIDIA Titan X up to the Challenges of Deep Learning?