Cray Products No Longer in Production
Cray XMT Massively Multithreaded Platform
The Cray XMT™ supercomputing system was a scalable massively multithreaded platform with a shared memory architecture for large-scale data analysis and data mining. It successor is the YarcData uRiKA graph appliance, a purpose built solution for Big Data relationship analytics.
Cray CX1000 Supercomputer
The Cray CX1000™ system was a dense, power efficient rack-mounted supercomputer that allowed customers to leverage Intel® Xeon® processors for scale-out cluster computing using dual-socket Intel Xeon 5600s (Cray CX1000-C), scale-through (GPU) computing leveraging NVIDIA Tesla® (Cray CX1000-G), and scale-up computing with SMP nodes built on Intel’s QuickPath Interconnect (QPI) technology offering "fat memory" nodes (Cray CX1000-S).
Cray CX1 Deskside Supercomputer
The Cray CX1™ was a deskside system configured for performance, functionality and cost, while appealing to a wide range of customers from the single user of a personal supercomputer to a department of users accessing shared clustered resources. The CX1-iWS™ system was a technical computing platform configured with Windows 7 and HPC Server 2008 R2, enabling engineers, scientists and researchers to run larger models and simulations.
Cray XT3, XT4, XT5 and XT6 Supercomputers
The Cray XT family of supercomputers was the first to break the production petascale performance barrier with over 2 petaflops on real-world applications on "Jaguar" – a Cray XT5™ system located at the National Center for Computational Sciences at Oak Ridge National Laboratory. The successor to the Cray XT is the Cray XE product family, built using the proven Cray XT infrastructure and incorporating AMD multicore processors and Cray's award-winning Gemini™ interconnect.
Cray X1E Supercomputer
The Cray X1E™ supercomputer combined the processor performance of traditional vector systems with the scalability of microprocessor-based architectures. High performance interconnect and memory subsystems allowed the Cray X1E system to scale from 16 to 8,192 processors, delivering up to 147 TFLOPS in a single system. The Cray X1E supercomputer and its predecessor, the Cray X1™ supercomputer, were the first vector systems designed to scale to thousands of processors in a single system image.
Cray XD1 Supercomputer
The Cray XD1™ supercomputer, introduced in 2004, was designed around a direct-connect processor approach to massively parallel processing that directly linked together processors, alleviating memory contention and interconnect bottlenecks found in cluster and SMP systems.
The Cray XD1 system embedded a high speed interconnect and application accelerators to remove major bottlenecks and improve performance on real-world applications. It was designed to meet the requirements of highly demanding HPC applications in fields ranging from product design to weather prediction to scientific research, to enable faster simulation and analysis, solve complex problems, and bring solutions to market sooner.
Cray SV1 and SV1ex™ Supercomputers
The Cray SV1 and Cray SV1ex vector supercomputing systems, introduced in 1998 and 2001, offered innovations in vector cache, MSP (Multi-Streaming Processor), and integrated SSD (Solid State Disk). Each system included a UNICOS operating system, software environment compatible with a variety of highly-tuned UNICOS applications, and CF90™ Fortran compiler that provided automatic vectorization, scalar optimization, and Autotasking™, a parallel-processing feature allowing programmers to partition individual programs so that the parts execute simultaneously on multiple processors.