Our vision is to put affordable, versatile, and easy to use and manage supercomputing resources directly in the hands of individuals, allowing them to get results faster and improve productivity of large-scale computing resources.

Beowulf clusters made it easy for everyone to acquire HPC resources, but using them productively remained hard. Rocketcalc makes HPC easy and productive for everyone.


Our partners include:

Our customers include:

Customer Testimonials:

“Rocketcalc makes really cool parallel machines that are affordable and convenient to use. Their customer service is excellent. The development and implementation of parallel algorithms has become so much easier. This has really helped our research.”

—Jan Verschelde, Associate Professor, University of Illinois at Chicago

“I couldn't be more pleased with my Rocketcalc cluster, or the support I've received from both Sami Mkaddam and Bryan Lewis. Both Sami and Bryan were the antithesis of hard-sell, and incredibly helpful in configuring and pricing a cluster that was within my budget, while ensuring it would meet my needs. I especially appreciated, upon delivery, how quickly we were up and running -- from unpacking, to plugging in, to adding users with the pre-configured system administration tools. It was a simple pleasure to receive a new computer system that ran out of the box, without any complications. The cluster remains busy practically 24x7 running experiments in parallel computational phylogeny (building evolutionary trees of life). The performance matches that of my other cluster (Mac OS X- based G5 Xserves), but at a fraction of the initial purchase cost. I wouldn't hesitate to recommend Rocketcalc LLC to anyone with the need for a turnkey personal computing cluster solution.”

—Marc L. Smith, Assistant Professor, Vassar College

“Rocketcalc's Titan cluster system has provided the Hydrologic Research Center (HRC) with a turn-key cluster solution contained in a space-efficient and well-integrated stand-alone design. The cluster management utilities were pre-installed on the front-end node and fully pre-configured with the clustered nodes for out-of-the-box operation upon arrival. Onsite installation required approximately 20 minutes. This alone is of great value for computational deployments where both staff-time investments and time-to-operation must be minimized to ensure that project deadlines are met on time and within budget. Matching this procurement value with the Titan cluster's proven success of running popular atmospheric, climate and ocean wave models (e.g., MM5 and Wavewatch III), Rocketcalc was an excellent choice for HRC's needs.

“Beyond the procurement and initial implementation, Rocketcalc's highly-responsive and personalized support services have added considerable value to the ongoing use of their cluster system. Their efficient expertise in supporting highly-technical configurations and inquiries is among the best we have experienced. Rocketcalc's excellent quality of service after the sale should be given special attention by anyone considering a clustered solution.”

—Jason Sperfslage, Systems Administrator / Software Engineer, Hydrologic Research Center San Diego, CA USA

“I have used Rocketcalc systems since 2004. The guys at Rocketcalc were able to custom build the cluster for my needs–both the hardware and software required by my research. When the system was delivered, Sami helped me get all the networking and software licenses working, and within a couple days I was compiling and running my code at awesome speeds. I have experience in building clusters piece-by-piece with individual boxes, and I will never do it again. Although the cost of the Rocketcalc was almost the same as purchasing the individual hardware for a self-build, the time saved with a turnkey system such as these is the largest benefit for me.”

—Patrick Lynett, Assistant Professor, Coastal and Ocean Engineering Division, Department of Civil Engineering, Texas A&M University

Bryan Blaine was a research scientist in a non-acoustics research group at the Naval Undersea Warfare Center in Newport RI before joining Rocketcalc. He also was involved in finite element code development as a consultant to the Advanced Materials Research Group at Virginia Tech. He holds a Master of Science degree in applied math.

Capabilities
Rocketcalc's affordable, compact cluster systems and easy-to-use cross-platform management software provide a unique platform for parallel applications. We can help embed applications in the cluster firmware and customize front-end management tools to produce a complete "cluster appliance" solution for Linux, Mac OS X and Windows workstations. Our systems are great for graphics, medical imaging, chemical modeling, engineering and many other application areas. Please contact us directly for discussing joint product development opportunities.