Better by Design & Built to Serve

CCSIservers.com continually at your service
     CCSIservers.com offers you the opportunity to leverage our expertise for all your server needs. You can generate quotes, place orders and review order history at your convenience on our new site.

CCSICustomers, leadership & Innovations
     Many high profile customers including AMD, Intel, HP, Universal Studios, the US Government, and many leading universities have trusted in CCSI’s expertise to provide innovative and reliable solutions. Among CCSI’s notable accomplishments in product design and manufacture are: The first IBM PC XT compatible Rackmount system for energy management control and telephony applications, the CCSI Imagination Station ™ was the 1st Multimedia authoring workstation, CCSI engineered and built one of the earliest rackmounted RAID storage servers in 1991 for Amblin division of Universal Studios, CCSI designed the 1 st 1U Dual slot 1 Pentium II Rack server for HPC clusters, CCSI developed the 1 st 1U Dual Athlon Rack Server, CCSI designed and manufactured the 1 st 1U rack server with redundant power supply, CCSI engineered one of 1 st green data center solutions, the CCSI Clusteron™, capable of running in 35C/95F ambient temperatures without additional air conditioning

AMD Magny-Cours Series Opteron Processors

AMD Opteron 6100 Press Release:
http://www.amd.com/us/press-releases/Pages/amd-sets-the-new-standard-29mar2010.aspx

Newly introduced AMD Opteron™ 6100 series processors, codenamed “Magny-Cours,” are available now from CCSIservers.com in a broad selection of Virtualization Servers, Storage Servers, Performance Servers, Clusters and HPC product lines. Eight Core and Twelve Core AMD Opteron™ systems are now available from CCSI in dual and quad processor 1U servers, 2U servers, 3U servers, 4U servers, and Direct Attached Storage (DAS) servers in 1U, 2U, 3U, 4U, 5U and 9U form factors. Custom design and configuration of servers and HPC clusters are also available from CCSI. Please contact one of our sales engineers for a free consultation.

Get more and do more for less when you go with AMD Opteron™ 6100 based servers. More CPU cores (choose 8 or 12 cores per CPU), more L3 cache (12MB per CPU) more DDR3 RAM capacity (up to 256GB in a 2 way system), more memory channels (4 vs. 3) plus you get 4P capable CPUs at no premium price with AMD! This means you can build a 48 core 4 way AMD Opteron™ 6100 series server for less you could have built a 24 core 4P server only a few months ago and it is not even an option with the competition!
Just in time for summer, AMD turns up the heat a few more notches in the technology race with the April 29, 2010 introduction of the Eight-Core and Twelve-Core Opteron™ processors code named “Magny-Cours.” Each of the eight or twelve cores is equipped with its own 128k of L1 cache plus 512k of L2 cache. There is also 12MB of L3 cache shared by all 8 or 12 cores. AMD Magny-Cours Opteron™ processors are multi processor (MP) capable, supporting up to 8 processors (with up to 96 cores) in one system and continue AMD’s evolutionary development of some of the most computationally powerful and yet energy efficient processors available. New performance enhancing features include: HyperTransportTM Technology 3 (HT3), Hyper Transport (HT) Assist, quad channel DDR3 memory at up to 1333MHz which doubles the memory capacity of the six-core Istanbul processors and AMD-V™ 2.0. Power saving features include: APML (Advanced Platform Management Link), comprised of the Remote Power Management Interface (RPMI) and the Precision Thermal Monitor; AMD CoolSpeed Technology, which reduces p-states when a temperature limit is reached; C1E, a deep sleep state, LV-DDR3 which helps to reduce overall power consumption.
CCSI, an early AMD Technology Partner with our advanced system cooling expertise has incorporated Eight-Core and Twelve-Core AMD Opteron™ processors in our entire AMD server product line. Coupled with the AMD 5600 Series chipset with I/O virtualization capability, HyperTransport™ 3.0 technology, and PCI Express® 2.0 the AMD Opteron™ 6000 product family is the ideal virtualization platform

Intel® Xeon® 5600 “Westmere” Series Processors

CCCSservers is pleased to announce the availability of the new Intel® Xeon® 5600 Series Processors (formerly codenamed "Westmere"). We've refreshed our entire line of  servers to take advantage of these new 32nm Nehalem micro-architecture four-core and six-core CPUs, offering the same performance as earlier Intel® Xeon® CPUs with 30% less power consumption. Features like Intel® Intelligent Power Technology, Intel® Turbo Boost Technology, Intel® Hyper-Threading Technology (Intel® HT Technology), and DDR3 memory, automatically regulate power consumption and increase processor frequency as needed and combine industry-leading energy efficiency with intelligent performance that adapts to your workload. Next generation Intel® Virtualization Technology enables best-in-class virtualization performance, superb scalability, enhanced flexibility, and simplified server management and Intel® AES (Advanced Encryption Standard) improves CPU performance as much as 52 percent for secure internet transactions.
Intel Xeon™ Nehalem microarchitecture processors in the 3000, 5000, 6500 and 7000 series, incorporate several breakthrough technologies that radically improve system speed and versatility. Intel® Turbo Boost Technology allows the system to automatically “shift into overdrive” by stepping up the clock speed as needed while the CPU(s) remain within their thermal envelope. Intel® QuickPath Interconnect (QPI) is their new integrated memory controller with a high-speed interconnect in each processor core, providing dynamically scalable interconnect bandwidth between processors and other components at up to 10.6GB/s with DDR3 1333. Hyper threading has been reintroduced in the Xeon™ processor 5500 series allowing more efficient use of the processor cores by means of thread-level parallelism. Intel® Intelligent Power Technology minimizes power consumption when server

components are not fully utilized. Automated low-power states automatically put processor and memory into the lowest available power states that will meet the requirements of the current workload. Integrated power gates allow idling cores to be reduced to near-zero power independently of other operating cores, reducing idle power consumption to 10 watts, versus 16 to 50 watts in prior-generations of Intel quad-core processors. Next-generation Intel® Virtualization

Technology enhances virtualization performance with new hardware-assisted capabilities across all elements of your server: Improvements to Intel Virtualization Technology (Intel VT-x) provide hardware-assisted page-table management, allowing the guest OS more direct access to the hardware and reducing compute-intensive

software translation from the VMM. Intel VT-x also

includes Intel VT FlexMigration and Intel VT Flex-Priority, which are capabilities for flexible workload migration and performance optimization across the full range of 32-bit and 64-bit operating environments. Intel Virtualization Technology for Directed I/O (Intel VT-d) helps speed data movement and eliminates much of the performance overhead by giving designated virtual machines their own dedicated I/O devices, thus reducing the overhead of the VMM in managing I/O traffic. Intel Virtualization Technology for Connectivity (Intel VT-c) further enhances server I/O solutions by integrating extensive hardwareassists into the I/O devices that are used to connect servers to the data center network, storage infrastructure and other external devices. By performing routing functions to and from virtual machines in dedicated network silicon, Intel VT-c speeds delivery and reduces the load on the VMM and server processors, providing up to 2x the throughput of non-hardware-assisted devices.

Performance Servers, Storage Servers, Blades, Clusters and HPC product lines are available now with Intel Xeon™ 5500 and 5600 series processors from CCSIservers.com in dual processor 1U servers, 1U and 2U twin servers, 2U servers, 3U servers, 4U servers, and Direct Attached Storage (DAS) servers in 1U, 2U, 3U, 4U, 5U and 9U form factors. Custom design and configuration of servers and HPC clusters are also available.  Please contact one of our knowledgeable sales engineers for a free consultation.

Awesome economy and massive storage capacity realized in Raidware™ servers. RAID, originally developed at UC Berkeley in 1988, and meaning Redundant Array of Inexpensive Disks, turned out to be a misnomer due to the expense of early RAID systems so the anagram was changed to mean Redundant Array of Independent Disks.  RAID was originally conceived to provide greater storage capacity but soon became popular as a means of providing increased reliability for data storage. Due to the overall cost of early RAID systems, they were originally used almost exclusively for mainframe storage.  Fast forward 21 years and 3.5” disk capacities are at 2TB and 2.5” disk capacities are at 500GB. As a result, RAID systems can be physically small while offering huge storage capacities, blazing performance and amazingly affordable prices. Now, packaged in shoe box sized devices with costs below $0.34/GB, RAID systems are becoming ubiquitous; finding use in video capture, editing and streaming, departmental storage, small office and home office (SOHO) storage and even in residential home network and media storage applications. Perhaps the anagram should be changed to RAVID for Redundant Array of Very Inexpensive Disks.

Stepping back in history, CCSI built the RapidRaid™, one of the earliest turnkey rack mounted storage server solutions, way back in 1991 for Universal Studio’s Amblin Division. The system was used to store the computer generated scenes outside the submarine on seaQuest DSV, a weekly television show from 1993-1996. The RapidRaid™, system featured temperature controlled cooling fans and twenty five hot swappable 1GB SCSI hard drives in a RAID 5 configuration providing an amazing [for its time] 21GB of usable storage space. The RapidRaid™, was so rugged it survived a Southern California earthquake without loss of data.  Times have changed, but CCSI still offers leading edge rack mounted storage solutions in configurations from 1U through 9U with raw storage capacities of 8TB to 100TB, as well as stand alone NAS RAID systems. Whatever your storage needs from a small departmental NAS to an iSCSI or fibre channel SAN to an industrial strength rack mounted storage solution, CCSI has your back. The 21st century question is not “can we afford RAID storage?” but rather, “can we afford NOT to utilize RAID storage?”

Everybody must get RAID!

Next Generation NVIDIA® Tesla 20-series GPUs, Code Named Fermi, Deliver Supercomputing Power at 1/10th the Cost

“One Researcher, One Supercomputer” is the goal of nVIDIA® through Tesla™ GPU technology. Compared to CPU-only systems, servers with Tesla 20-series GPU Computing Modules deliver supercomputing power at 1/10th the cost and 1/20th the power consumption of common quad core CPU solutions. Cluster level supercomputing power is available and affordable right at your desk and powered from a standard power strip! The NVIDIA® Tesla™ Personal Supercomputer is based on the revolutionary NVIDIA® CUDA™ parallel computing architecture and powered by up to 512 parallel processing cores per GPU module. These amazingly powerful stand alone computers cost in the thousands rather than the hundreds of thousands of dollars, easily putting them within reach of scientists and engineering professionals.
The NVIDIA® Tesla™ 20-series GPUs, based on the next generation CUDA GPU architecture codenamed “Fermi”, make GPU and CPU co-processing pervasive by addressing the full-spectrum of computing applications. Designed for C++ and available with a Visual Studio development environment, it makes parallel programming easier and accelerates performance on a wider array of applications than ever before – including dramatic performance acceleration in ray tracing, physics, finite element analysis, high-precision scientific computing, sparse linear algebra, sorting, and search algorithms.
Fermi features several major innovations:
• 512 CUDA cores
• NVIDIA Parallel DataCache technology
• NVIDIA GigaThread™ engine
• ECC support
These features provide 8X the double precision performance compared Tesla 10-series GPU computing products. When compared to the latest quad-core CPU, Tesla 20-series GPU computing processors deliver equivalent performance at 1/20th the power consumption and 1/10th the cost. NVIDIA Tesla 20-Series GPUs (Fermi architecture) are now available from CCSIservers.com. Choose a GPU-powered 1U server with Intel Xeon 5500 or 5600 CPUs or with AMD Opteron 6100 processors, or consider a personal supercomputer with up to 3 20-Series Tesla cards or a 4U rack mounted server with up to 8 20-Series Tesla cards.
Affordably accelerate your research with GPU Supercomputing servers and workstations, in tower or rack mounted configurations using Nvidia® Tesla™ Fermi GPU modules with up to two GPUs in a 1U rack system, up to four GPUs in a tower, or up to eight GPUs in 4U rack system; netting better performance than a small computer cluster, with 2 Teraflops in a 1U rackmount, 4 Teraflops in tower server and 8 Teraflops in a 4U rackmount server. These GPU optimized solutions are ideal for medical imaging, genomics, biometrics, financial simulation, electrodynamics, computational fluid dynamics, 3-D rendering, computational chemistry modeling, protein folding, EDA, oil & gas exploration, aerospace, astrophysical, military simulation, and other computationally intensive applications.

CCSI is one of the oldest custom computer manufacturers. Beginning with a rack-mountedcomputer controlled energy management and cogeneration installation at Candler General Hospital in 1986 and progressing on to rackmounted payphone call processing systems in 1987, CCSI has been an innovator and