Fact-checked by Grok 2 weeks ago

Cray Y-MP

The Cray Y-MP is a vector-processing series developed by Research, Inc., and introduced in 1988 as a direct successor to the , featuring scalable with up to eight central processing units (CPUs) clocked at 166.7 MHz (6-nanosecond cycle time) and up to 64 million 64-bit words (512 MB) of central memory. It was the world's first supercomputer to sustain over one gigaflop (billion floating-point operations per second) of performance in practical workloads, marking a significant advancement in for scientific simulations and applications. The Y-MP architecture built on the vector register design pioneered in earlier systems, incorporating very-large-scale integration (VLSI) chips for improved efficiency and density, while supporting a 32-bit and optional solid-state disk (SSD) storage capacities ranging from 128 million to 1 billion words for high-speed data access. Configurations like the Y-MP8/864 model, installed at sites such as the Ohio Supercomputer Center in for $22 million, achieved peak speeds of approximately 2.7 gigaflops and were recognized as the fastest supercomputers available at the time, enabling real-time processing of complex problems in fields like and modeling. Systems ran primarily on the UNICOS operating system, a UNIX-based environment with vectorizing compilers, and maintained with the older system for up to four processors and 16 million words of memory. The Y-MP series represented a pivotal in during the late 1980s and early 1990s, powering major research installations such as the National Center for Atmospheric Research's Shavano system from 1990 to 1997, where it sustained over one gigaflop on ocean models and set performance benchmarks before being superseded by more advanced designs like the C90. Variants including the air-cooled Y-MP EL (using technology with 30-ns cycles and up to four CPUs) and the Y-MP2E (limited to one or two processors with up to 128 million words of memory) expanded accessibility for mid-sized computing needs, while emphasizing scalability, reliability, and power efficiency without requiring specialized motor generators. Overall, the Y-MP solidified Research's dominance in the market, influencing decades of innovations through its balance of raw speed, , and I/O capabilities.

History

Development Background

The Cray Y-MP emerged as the direct successor to the supercomputer, which had been introduced in , with the primary goal of addressing the intensifying computational requirements of scientific simulations through enhanced support for greater numbers of processors, accelerated processing rates, and expanded memory resources. This evolution was driven by the need to sustain Cray Research's dominance in amid surging demands from applications requiring massive parallel operations. Development of the Y-MP was spearheaded by principal designer at Cray Research, who had previously led the X-MP project; Chen initiated the Y-MP effort before departing the company in 1987 to pursue independent ventures backed by , after which the design was finalized by a team under executive vice president Lester T. Davis. This work unfolded prior to founder Seymour Cray's exit from Cray Research in 1989 to establish his own firm, Cray Computer Corporation. Conceptualization began in the mid-1980s as an extension of the X-MP's multiprocessor framework, focusing on engineering decisions that prioritized reliability, scalability, and compatibility with existing software ecosystems. A pivotal innovation was the integration of very large-scale integration (VLSI) (ECL) technology, featuring custom 2500-gate ECL gate arrays that enabled denser circuit packing and substantially fewer discrete components per processor than in the X-MP, thereby improving manufacturing efficiency and system reliability. These advancements were informed by balanced scalar and vector processing architectures, along with multiport memory designs to optimize data access in multiprocessor environments. The Y-MP's engineering emphasized multi-processor scalability tailored to vector workloads, partly as a strategic response to rising competition from Japanese firms like and , whose vector systems were gaining traction in global markets. This focus aligned with critical applications in , weather modeling, and , where the need for sustained high-throughput computations drove the push for architectural refinements over the X-MP baseline.

Release Timeline

The Cray Y-MP series began shipping in 1988, with the first system delivered to in the fall of that year. Early adopters included U.S. national laboratories such as , which obtained six Y-MP systems, and , which accepted its initial unit in 1989. The rollout was facilitated by the availability of the UNICOS operating system, which supported the Y-MP from its debut and ensured compatibility with prior Cray architectures. In 1990, Cray Research expanded the lineup with the Model E, which improved scalability for larger computational environments compared to the original Model D. This evolution addressed growing demands for multi-processor scalability in scientific and applications. The series continued to diversify in 1992 with the introduction of the Y-MP M90, a variant that employed memory to lower costs while enabling expanded capacities for data-intensive workloads. Concurrently, the Y-MP EL debuted as an entry-level, air-cooled model designed to broaden beyond high-end facilities. Cray Research manufactured more than 200 Y-MP systems overall before shifting production to the C90 successor. Priced from roughly $15 million to $30 million per system depending on configuration, the Y-MP achieved peak market adoption in the early , prior to industry transitions toward technologies.

Architecture

Processor Design

The Cray Y-MP features a multiprocessor with up to eight identical , each serving as a (CPU) capable of handling scalar, , and generation tasks. Each operates on a 6-nanosecond clock cycle, equivalent to 167 MHz, and employs bipolar (ECL) implemented via high-density very-large-scale integration (VLSI) chips to achieve high-speed operations. The core of each includes two primary functional units for arithmetic: an add pack and a pack, enabling parallel floating-point operations on 64-bit data. These units support pipelined execution, allowing two results per clock cycle in optimal conditions. Vector processing in the Y-MP is designed for high-throughput scientific computations, utilizing vectors of up to elements stored in dedicated vector registers. Operations can employ mode, which links multiple functional units to process dependent vector instructions without intermediate , thereby sustaining peak throughput by allowing continuous flow between units. Complementing this, choke mode permits selective pausing of vector execution to synchronize with dependencies or memory access, enhancing flexibility in complex algorithms. The absence of a traditional is offset by reliance on these vector registers and chaining mechanisms, which minimize by keeping in fast on-chip pipelines rather than requiring frequent memory fetches. The scalar processing component within each CPU comprises seven functional units: dedicated units for integer addition, logical operations, shifting, and population/parity counting, alongside shared floating-point units for addition, multiplication, and reciprocal approximation that serve both scalar and modes. Address generation is handled by separate 32-bit add and multiply units, supporting extended addressing modes. This configuration yields a theoretical peak performance of 333 MFLOPS per , derived from two floating-point operations per 6 cycle. In multiprocessor configurations, the Y-MP employs a uniform shared-memory model where all processors access a common central through a high-bandwidth , facilitating inter-processor communication and without dedicated interconnect hierarchies. This design promotes while maintaining coherent access to up to 1 GB of static . Compared to its predecessor, the , the Y-MP doubles the clock speed from 9.5 ns to 6 ns and integrates VLSI ECL chips, which reduce overall power consumption and cooling requirements—though it retains liquid with —enabling denser packaging and higher reliability in multi-processor setups.

Memory System

The central memory of the Cray Y-MP Models D and E consisted of static RAM (), configurable in capacities up to 1 and organized into 64-bit words with an 8-bit SECDED error-correcting code for single-error correction and double-error detection. This memory was 256-way interleaved in the 8-processor Y-MP8 configuration (with fewer banks in lower models, such as 128 for Y-MP4 and 64 for Y-MP2) to facilitate concurrent accesses from multiple processors and I/O channels, achieving a memory device access time of 15 ns. The base hardware design omitted native support, relying instead on the UNICOS operating system's paging mechanisms to enable virtual addressing for larger workloads. Pipelined memory access protocols sustained high throughput, with each processor featuring four independent ports capable of 4-word (32-byte) bursts per clock cycle to match vector unit demands. In an 8-processor system, this design supports high aggregate bandwidth sufficient for parallel vector computations without significant contention under balanced loads. Subsequent variants prioritized capacity and affordability over speed by adopting dynamic RAM (). The Y-MP M90 replaced SRAM with DRAM offering up to 32 , albeit with a slower 50 ns access time, enabling cost-effective handling of memory-bound simulations and databases. Similarly, the Y-MP EL provided 32 MB to 1 of DRAM in configurations with reduced interleaving (e.g., 32- or 64-way), lowering system costs for mid-range scientific computing while preserving compatibility with Y-MP vector pipelines.

I/O Subsystem

The I/O Subsystem () of the Cray Y-MP serves as the primary interface for data movement between the mainframe's central memory, front-end processors, peripheral devices, and secondary storage, enabling efficient () operations without burdening the vector processors. It employs dedicated I/O Processors () as scalar units to orchestrate these transfers, supporting high-throughput workloads in scientific computing environments. The IOS incorporates up to eight in the Model E configuration, each functioning as an independent 16-bit scalar with 64 Kparcels of local for task-specific operations like and handling. These manage transfers at rates of 100 MB/s per channel to central , ensuring low-latency peripheral access. In contrast, the Model D supports up to four , providing scalable functionality for various installations. Common IOP variants include the IOP for front-end networking, Buffer IOP for temporary staging, and Disk IOP for storage device control. Channel architecture in the IOS supports diverse connectivity through up to 64 front-end and rear-door channels, facilitating simultaneous transfers from devices like drives and disks under the operating system's drivers. High-speed options include HIPPI interfaces at 100 MB/s for data links and early Ethernet-compatible channels at lower rates for gateways, with aggregate capacities reaching 3.2 /s in multi-IOP arrangements. A shared 4 Mword buffer (expandable to 32 Mwords) temporarily holds en route to central memory, minimizing bottlenecks during intensive I/O phases. The optional Solid-State Disk (SSD) integrates as high-speed non-volatile storage for staging large simulation datasets, available in capacities from 512 MB to 4 GB using modules with single-error correction and double-error detection. Connected via dedicated IOP pairs, the SSD sustains rates of MB/s in 64-word blocks, accelerating access times compared to magnetic media. IOPs and channel electronics are liquid-cooled via the mainframe's refrigeration system, including heat exchangers and dielectric fluid circulation to dissipate thermal loads from high-bandwidth operations, ensuring reliable integration within the Y-shaped layout for multi-processor models.

Models and Variants

Model D

The Cray Y-MP Model D, introduced in 1988, represented the foundational configuration of the Y-MP series, built on (ECL) technology for high-performance vector processing. It featured a liquid-cooled mainframe , available in single-bay or multi-bay designs to accommodate varying scales of subsystems and . This design prioritized computational speed for demanding scientific workloads, with an 8-processor variant drawing approximately 150 kW of power. Available in Y-MP/2D, /4D, and /8D variants, the Model D supported 2 to 8 processors, enabling scalable parallelism within a compact . Memory consisted of 256 to 512 of static () as standard, expandable to 1 optionally, reflecting a deliberate focus on low-latency access over larger capacities; notably, it offered no support for dynamic (). These specifications positioned the Model D as the baseline system for the Y-MP lineup, optimized for high-end research environments requiring rapid vector operations. Production of the Model D commenced in 1988, with the initial deployments exceeding 20 units by 1990, directed primarily toward laboratories for advanced computational tasks. Overall, approximately 200 Y-MP systems, including Model D configurations, were manufactured through 1994.

Model E

The Cray Y-MP Model E, released in 1990 as the successor to the Model D, represented an enhancement in for the Y-MP series, enabling configurations from 2 to 8 processors in multi-cabinet setups. These systems utilized (ECL) processors similar to earlier Y-MP designs, but with expanded support for larger workloads through improved memory and I/O capabilities. The Model E was produced until 1994, with approximately 200 units of the Y-MP series manufactured overall, many of which were Model E variants deployed in demanding scientific computing environments. Key configurations ranged from the Y-MP/2E (1-2 CPUs) to the Y-MP/8E (8 CPUs across two cabinets), with the Y-MP/8I offering a single-cabinet integrated option for 4-8 CPUs. Central memory capacity scaled up to 256 million words (2 ) of static () in the largest setups, such as the Y-MP/8E, providing better support for memory-intensive applications compared to prior models. The I/O subsystem was significantly broadened, supporting up to 8 processors () in the Y-MP/8E, which facilitated higher for peripheral devices and data transfer in multi-processor environments. These systems could span up to four cabinets when including additional I/O and storage bays, enhancing overall system modularity. A notable feature was the expanded solid-state disk (SSD) support using Model E technology, configurable up to 512 million words (4 ), which improved throughput for large-scale and job in scientific simulations. Designed for extended scientific projects, the Model E series cost between $20 million and $30 million depending on configuration, reflecting its positioning as a high-end vector supercomputer for institutions handling complex computational tasks through the mid-1990s.
ConfigurationCPUsCentral Memory (Mwords)IOPsSSD Options (Mwords)Cabinets
Y-MP/2E1-232-641-2128-5121
Y-MP/4E1-464-1281-4128-5121-2
Y-MP/8E8128-2561-8128-5122
Y-MP/8I4-864-1281-4128-5121

Y-MP M90

The Cray Y-MP M90 represented a memory-optimized evolution of the Y-MP series, introduced in as a response to demands for greater main in environments. Unlike the standard Y-MP models that relied on static RAM (), the M90 employed dynamic RAM () technology, enabling configurations with up to 8 processors and memory capacities ranging from 8 to 32 —approximately four times the maximum of earlier variants. This shift to high-density 4 Mbit or 16 Mbit chips allowed for more compact and cost-effective memory subsystems while preserving the core vector-processing of the Y-MP. Key specifications of the Y-MP M90 included a memory access of around 50 ns, slower than the 21 ns provided by in standard Y-MP systems, but this trade-off delivered substantially higher capacity at a reduced cost per byte, making large datasets more accessible without prohibitive expense. The system maintained full compatibility with the UNICOS operating system and existing Y-MP , allowing users to leverage prior investments in code and tools without modification. With up to 17.1 GB/s of aggregate across four ports per processor, the M90 supported efficient data movement for vectorized workloads, though its design emphasized capacity over the raw speed of SRAM-based predecessors. The Y-MP M90 was particularly targeted at memory-bound applications, such as large-scale scientific modeling and simulations where dataset sizes exceeded the limits of conventional supercomputers, enabling computations that were impractical on smaller-memory systems. A single-cabinet configuration option enhanced its space efficiency, housing the full CPU, memory, and I/O components in one unit for installations with constrained footprints. Production of the M90 was limited, with an estimated run of around 50 units, serving as a transitional model that bridged the Y-MP era to the subsequent C90 series by demonstrating the viability of in high-end vector supercomputing.

Y-MP EL

The Cray Y-MP EL was introduced in 1992 as an entry-level supercomputer designed to provide Cray's vector processing capabilities at a more accessible price point, targeting universities and smaller research laboratories. It utilized CMOS logic technology to enable air cooling without the need for freon refrigeration, significantly reducing power consumption and installation complexity compared to higher-end liquid-cooled models. This air-cooled design allowed the system to fit within a single compact cabinet, making it suitable for environments with limited space and infrastructure. Configurations of the Y-MP EL supported 1 to 4 processors, with memory ranging from 32 to 1 of . Each operated at a 30 ns clock period (approximately 33 MHz), delivering a peak performance of 133 MFLOPS per CPU—substantially lower than the mainline Y-MP's 166 MHz clock and 333 MFLOPS due to the slower implementation. The system maintained full binary compatibility with the UNICOS operating system and software from earlier Cray architectures, inheriting their scalar and processing heritage while prioritizing affordability over peak throughput. Priced starting at around $340,000 for a two-processor , the Y-MP EL aimed to broaden Cray's market beyond supercomputing installations, with approximately 130 units ordered in its initial year. Its reduced performance and entry-level focus made it ideal for computational tasks in academic and mid-tier industrial settings, such as and automotive research, without requiring the extensive support infrastructure of flagship systems.

Performance

Benchmark Results

The Cray Y-MP's performance was rigorously evaluated using standard benchmarks, with the Linpack test emerging as a key metric for floating-point capabilities. In 1989, an 8-processor configuration achieved a sustained performance of 2.144 GFLOPS on the optimized Linpack benchmark, marking it as the first to surpass the 1 GFLOPS barrier for sustained operations across multiple applications. This result utilized 64-bit floating-point precision and demonstrated an efficiency of approximately 80% relative to the system's theoretical peak of 2.667 GFLOPS, enabled by the processor's functional unit chaining that allowed overlapping of arithmetic operations in vector pipelines. Benchmark testing on the Y-MP primarily involved vector triad kernels, such as those in the Linpack suite (e.g., DAXPY operations), to measure rates under vectorized conditions. These kernels stressed the system's registers and pipelines, revealing high efficiency on codes that maximized —where the output of one functional unit could immediately feed into another without stalls. For real-world workloads, evaluations extended to kernel benchmarks like the Livermore Loops, which simulated scientific tasks and confirmed sustained rates approaching 2 GFLOPS on optimized applications. Within the Y-MP family, scaled with count and model variants. A single- Y-MP achieved 0.324 GFLOPS sustained, while scaling to eight processors yielded near-linear gains up to 2.144 GFLOPS, highlighting the system's effective shared-memory .
ConfigurationProcessorsClock (ns)Sustained Linpack (GFLOPS) (GFLOPS) (%)
Y-MP/832862.1442.667~80
Y-MP/832461.1591.333~87
Y-MP/832160.3240.333~97
Y-MP M98861.7332.666~65
Pre-TOP500 performance lists, based on Linpack and kernel results compiled by researchers like Jack Dongarra, consistently ranked the Y-MP as the world's fastest supercomputer from its 1988 debut through 1993, before massively parallel systems like the CM-5 overtook it.

Comparative Analysis

The Cray Y-MP represented a significant advancement over its predecessor, the Cray X-MP, delivering approximately three times the peak performance at 2.667 GFLOPS compared to the X-MP's 800 MFLOPS across four processors. This improvement stemmed from faster clock speeds and enhanced vector processing efficiency, with the Y-MP achieving up to 52% higher vector MFLOPS rates per processor. Additionally, the Y-MP offered up to four times the central memory capacity of the X-MP, supporting configurations of 512 MB versus the X-MP's maximum of 128 MB, which improved multi-processor balance and reduced contention in shared-memory operations. In comparison to contemporary Japanese supercomputers, the Y-MP demonstrated superior vector throughput in multi-processor configurations, sustaining over 1 GFLOPS on applications where single-processor limits constrained competitors like the VP-2600 (peak 5 GFLOPS but asymptotic SAXPY performance around 1.1 GFLOPS) and the SX-2 (peak 1.3 GFLOPS). For instance, on vectorized benchmarks such as Pueblo64, the Y-MP's eight-processor setup outperformed the VP-2600 by leveraging better load balancing, though the VP-2600 excelled in isolated contiguous memory tasks by factors of 4 to 9. Similarly, the Y-MP's shared-memory provided more consistent throughput than the SX-2's pipelined design on longer vectors, where the SX-2 achieved 2 to 4 times the speed of the earlier X-MP but lagged behind the Y-MP in scaled workloads. On the list, introduced in , Y-MP systems frequently ranked in the top 10 with Linpack scores up to about 2.1 GFLOPS, maintaining leadership among vector processors until systems like the C-240 emerged with optimized RISC-based around . Key metrics highlighted the Y-MP's efficiency, including a cost-performance of approximately $10 million per GFLOPS, based on system prices around $13 to $40 million for configurations delivering 2 to 2.67 GFLOPS . Its shared-memory model offered scalability advantages over massively parallel processing () alternatives like the , enabling easier programming for scientific codes with uniform memory access and up to 8-way multiprocessing without the communication overhead of distributed-memory systems, which limited Paragon efficiency to 20-30% on irregular workloads. Despite these strengths, the Y-MP's limitations included high power consumption of around 120-200 kW for eight-processor models, necessitating extensive liquid cooling infrastructure that increased operational costs compared to emerging RISC-based clusters in the early 1990s, which achieved similar throughput at under 50 kW per GFLOPS.

Applications and Deployments

Scientific Computing Uses

The Cray Y-MP supercomputer played a pivotal role in advancing computational fluid dynamics (CFD) simulations, particularly for aerospace applications at NASA, where its vector processing capabilities enabled high-fidelity modeling of complex flows such as those encountered in aircraft design and propulsion systems. Researchers at NASA's Numerical Aerodynamic Simulation (NAS) facility utilized the Y-MP to execute Navier-Stokes equations for viscous flow simulations, including chemistry-involved reacting flows and supersonic problems, achieving significant reductions in computation time for iterative solvers. In climate modeling, the National Center for Atmospheric Research (NCAR) leveraged the Y-MP for weather and global climate simulations, processing large-scale atmospheric data to predict patterns and long-term trends with enhanced resolution. These efforts included running vectorized general circulation models like CCM2 on the Y-MP, which supported ensemble simulations for medium-range forecasting and ocean-atmosphere coupling studies. For nuclear weapons design at Lawrence Livermore National Laboratory (LLNL), the Y-MP facilitated shock physics computations essential for simulating hydrodynamic behaviors under extreme conditions, aiding in the certification of weapon performance without physical testing. Such applications relied on the system's ability to handle dense linear algebra operations, including Gaussian elimination methods for solving seismic wave propagation equations in boundary element models, which improved accuracy in geophysical analysis by managing large sparse matrices efficiently. In , the Y-MP's (FFT) implementations were instrumental for in N-body simulations of formation and cosmological structure evolution, where high-performance FFT libraries processed spectral data from gravitational dynamics models. The system's software ecosystem, including optimized vectorized libraries such as the Cray Scientific Library (SCILIB) and FFTPAK for mathematical routines, automated parallelization and vector chaining to exploit the Y-MP's multiple processors. Under the UNICOS operating system, tools like the Network Queuing System (NQS) enabled efficient batch job scheduling for long-running simulations, allowing multiple users to submit and manage vector-intensive workloads in a shared environment. The Y-MP supported datasets up to approximately 512 MB in central memory, augmented by for staging larger files during I/O operations, which permitted simulations involving grid sizes exceeding those feasible on the predecessor X-MP by factors of 2 to 3 in effective throughput per processor. This scale enabled significant accelerations in overall simulation turnaround for multi-processor configurations compared to X-MP-era systems, transforming the feasibility of grand-challenge problems in these domains.

Major Installations

The Cray Y-MP saw prominent deployments at several key research institutions, beginning with its first installation at Ames Research Center in fall 1988, where it replaced one of two systems and supported the Numerical Aerodynamic Simulation (NAS) program for (CFD) applications. An additional eight-processor Y-MP/832 configuration was installed at Ames in fall 1989, replacing a /48 and further advancing aerospace simulations. The Pittsburgh Supercomputing Center received an eight-processor Y-MP/832 in fall 1988 (operational from 1989 to 1993), replacing its Cray X-MP/48 and enabling advanced materials science research as well as broader scientific computing workloads. Similarly, the Los Alamos National Laboratory ordered Y-MP systems in summer 1988, with installations supporting nuclear weapons simulations and serving as a precursor to the Accelerated Strategic Computing Initiative (ASCI) through early parallel processing efforts in the early 1990s. The Ohio Supercomputer Center installed an eight-processor Y-MP8/864 in August 1989, which was recognized as one of the fastest supercomputers at the time and supported a wide range of scientific simulations. The (NCAR) deployed the Y-MP8 system named "Shavano" in 1990, serving as its flagship until 1997 for and atmospheric modeling, and a Y-MP2 system named "" in 1991 dedicated to additional simulations. Other notable sites included the UK Meteorological Office, which installed an eight-processor Y-MP with in winter 1990 to enhance models, including early global circulation simulations. The University of Texas System's Center for acquired a Y-MP8 in spring 1991, alongside an earlier Y-MP2/116 at Texas A&M in fall 1989, facilitating academic research in various scientific domains. By the mid-1990s, the Y-MP had been widely adopted, with systems in operation at over 100 sites worldwide, though many were retired in favor of newer massively parallel architectures like the Cray T3D (introduced 1993) and T3E (1996). These deployments enabled key breakthroughs, such as improved resolution in global circulation models for weather and climate prediction at sites like the Meteorological and NCAR.

Legacy

Technological Influence

The Cray Y-MP supercomputer significantly influenced subsequent designs within Cray Research's product line, serving as the architectural foundation for the C90 system introduced in 1991, which expanded to 16 processors and achieved a peak performance of 16 GFLOPS through enhancements like larger 10,000-gate ECL arrays from and doubled vector pipes per CPU. This evolution built directly on the Y-MP's scalable processing and shared-memory model, enabling higher parallelism while maintaining binary compatibility with existing UNICOS software. The Y-MP's design also extended to -based successors, with the Y-MP EL variant—implementing the core Y-MP architecture in more cost-effective technology—paving the way for the J90 series minisupercomputer produced from 1994 to 1998. Furthermore, the Y-MP's emphasis on tightly coupled shared-memory paradigms carried forward into broader industry developments, notably influencing the 1996 merger between Research and Inc. (SGI), where SGI's expertise complemented 's high-performance shared-memory systems to unify scalable architectures. In terms of industry shifts, the Y-MP accelerated the adoption of very-large-scale integration (VLSI) techniques in by employing high-density (ECL) VLSI for its processors, which supported faster scalar and operations compared to prior discrete-component designs. It also contributed to the ongoing development of the UNICOS operating system, a Unix variant that originated with earlier systems and evolved into the modular UNICOS/mk kernel for massively parallel processing (MPP) machines like the T3E in the mid-1990s. The Y-MP helped establish early standards for gigaflops-scale computing, with multi-processor configurations delivering sustained performance exceeding 2 GFLOPS, which became a benchmark for vector supercomputers entering the 1990s HPC expansion. This positioned Cray Research at the forefront of the decade's supercomputing growth, training engineers and users in advanced parallel programming that supported the broader HPC workforce boom. Economically, the Y-MP's commercial success drove Cray Research's revenues to a peak of $922 million in 1994, reflecting strong demand before competitive pressures from MPP alternatives intensified.

Cultural References

The Cray Y-MP has appeared in several notable depictions within film and television, underscoring its role as an icon of cutting-edge computing in the early . In the 1992 thriller , directed by , the supercomputer is prominently featured as a powerful cryptography-breaking machine central to the plot, where a team of hackers infiltrates a facility housing the system to access a black-box decryption device. The film's portrayal highlighted the Y-MP's immense computational power, reflecting contemporary perceptions of supercomputers as tools for and . Beyond , the Y-MP has been referenced in various forms of that chronicle the evolution of . It features in documentaries exploring the 1980s computing revolution, such as discussions of Cray Research's innovations in vector processing and parallel architectures that propelled scientific advancements. In technical literature, the system is often cited as a pivotal milestone, symbolizing the era's push toward gigaflop-scale performance; for instance, its achievement of sustained 1 GFLOPS in 1988 marked the first time a reached this threshold on real-world applications, fueling narratives about the dawn of tera-scale computing. The Y-MP contributed significantly to public fascination with supercomputing, embedding "" as a for computational machinery in popular discourse. News coverage of its 1988-1989 performance milestones, including the 1 GFLOPS barrier, amplified its mystique in science fiction and , portraying it as a harbinger of technological supremacy amid Cold War-era computing races. In contemporary contexts, preserved Y-MP systems serve as educational artifacts, evoking the era's engineering feats. A Cray Y-MP EL model from 1992 is on exhibit at the in , allowing visitors to explore its liquid-cooled design and historical significance in the supercomputing lineage.

References

  1. [1]
    [PDF] The Cray Series of Supercomputers - Edward Bosworth
    The Y–MP was introduced in 1988, with up to eight processors that used VLSI chips. It had a 32–bit address space, with up to 64 megawords of static RAM main ...
  2. [2]
    CRI Cray Y-MP8 - Shavano - cisl.ucar.edu
    Shavano, a Cray Y-MP8 supercomputer, was used from 1990-1997, had 8 processors, 166.7 MHz clock speed, and was the top of the line for its time.
  3. [3]
    [PDF] Introducing the Cray Y-MP2E Supercomputer, 1990
    The CRAY Y-MP2E system is available in one- and two- processor configurations with up to 64 million words of central memory. The CPU - the same as that used on.
  4. [4]
    Cray History - Supercomputers Inspired by Curiosity - Seymour Cray
    TECH STORY: The Cray Y-MP was the world's first supercomputer to sustain over 1 gigaflops. Considered a follow-on to the X-MP, the initial system had eight ...
  5. [5]
    Cray Y-MP8/864 - Ohio Supercomputer Center
    The $22 million Cray Y-MP8/864 system, which was deemed the largest and fastest supercomputer in the world for a short time.
  6. [6]
    [PDF] The Cray Extended Architecture Series of Computer Systems, 1988
    The EA series is based on the CRAY YMP architecture and builds upon the best features of the field-proven. CRAY X-MP series while providing customers with.<|control11|><|separator|>
  7. [7]
    Dr. Steve Chen - IT History Society
    He is best known as the principal designer of the Cray X-MP and Cray Y-MP multiprocessor supercomputers. He left Cray Research in 1987. With IBM's financial ...
  8. [8]
    The Next Generation at Cray - The New York Times
    Feb 10, 1988 · Chen started the Y-MP project, he left it in 1985, and the design was completed by a team headed by Lester T. Davis, an executive vice ...Missing: development | Show results with:development
  9. [9]
    [PDF] The CRAY-MP Series of Computer Systems
    The CRAY Y-MP series may also be configured with an optional SSD solid-state storage device of 32-, 128-,. 256-, or 512-million word capacity to address even.
  10. [10]
    Cray Y-MP, decomissioned | PSC
    The Y-MP retained software compatibility with the X-MP, but extended the address registers from 24 to 32 bits. High-density VLSI ECL technology was used and a ...Missing: history | Show results with:history
  11. [11]
    Cray Channels Extract – Customer notes - Cray-History.net
    The CRAY Y-MP system replaces a CRAY X-MP system that was installed in 1987. ... CRAY Y-MP and CRAY X-MP systems. 31. MCC, Cray-SuperServer, Volume 16, Number 2 ...
  12. [12]
    [PDF] Windows on Computing | Los Alamos National Laboratory
    The Laboratory obtains the first of its six Cray Y-MP com- puters. It also in- stalls, studies, and evaluates a number of massively parallel computers. The ...Missing: adopters | Show results with:adopters
  13. [13]
    Cray Chips Newsletter - Cray-History.net
    CRAY_CHIPS-1989-04-13, CRAY Y-MP SYSTEM ACCEPTED,Serial 1007 at the Lawrence Livermore National Laboratory in California passed all tests and was accepted on ...Missing: adopters | Show results with:adopters
  14. [14]
    Cray YMP Model E - Cray-History.net
    The Cray-YMP Model E range was the direct successor to the YMP Model D and were manufactured from 1990 until 1994. Around 200 YMP machines were manufactured.
  15. [15]
    CRAY LAUNCHES MASSIVE MEMORY Y-MP M90s, CUTS LOW ...
    May 14, 1992 · Eagan, Minnesota-based Cray Research Inc has introduced a new Y-MP M90 series of supercomputers, claiming that they offer the largest main ...
  16. [16]
    [PDF] CRAY Y-MP~ Computer Systems Functional Description Manual
    Three additional functional units are used for both scalar and vector ... The Scalar Shift functional unit and Vector Shift functional unit shift 64-bit ...Missing: choke | Show results with:choke
  17. [17]
    [PDF] The CRAY X-MP Series of Computers, 1983
    The CRAY X-MP Computer System, with its 9.5 nsec clock cycle time, is the fastest general-purpose computer system commercially available today. The X-MP is ...Missing: ns | Show results with:ns
  18. [18]
    The Cray Y-MP-a VLSI supercomputer - IEEE Computer Society
    At a 6-nsecclock cycle, the Y-MP offers 30 times theprocessing power of the original Cray-1.The main memory in the Y-MP is 32 million64 bit words, arranged ...
  19. [19]
    [PDF] The CRAY Y-MP M90 supercomputer series - Bitsavers.org
    In a single cabinet, the CRAY Y-MP M94 system offers up to four CPUs and 16 Gbytes. (2 Gwords) of central memory. With outstanding price/performance at a low ...
  20. [20]
    [PDF] CRAY Y-MP EL Supercomputer System
    CRAY Y-MP EL Product Specifications. CPU. Technology. Clock period. Number of CPUs. CMOS. 30 ns. 1-4. Peak performance (per CPU) 133 MFLOPS. Memory. Memory ...
  21. [21]
    [PDF] CRAY Y-MP C90 Functional.Description Manual - Bitsavers.org
    Section 3, "I/O Subsystem," describes the basic architecture and functions of the input/output subsystem (IDS). A specification sheet is included at the end ...
  22. [22]
    Cray YMP Model D - Cray-History.net
    The Cray-YMP runs on UNICOS a UNIX operating system. The CRAY Y-MP CPUs and memory, one I/O Subsystem and the SSO are connected in a Y shaped configuration.Missing: DI/ | Show results with:DI/
  23. [23]
    Cray FAQ Part 5: Machine Specifications - 0x07Bell.net
    Cray YMP Model E IOS​​ Second generation YMP Vector Supercomputer. Available with 128,256 or 512 Mword SSD.
  24. [24]
    [PDF] The CRAY Y-MP C90 Supercomputer System
    The CRAY Y-MP C90 is a powerful supercomputer with 16 processors, 16 GFLOPS peak performance, 64-way vector parallelism, and 250 Gbytes/sec memory bandwidth.
  25. [25]
    [PDF] ON PARALLEL COMPUTERS - ECMWF
    The Cray Y-MP has a peak performance (with perfect vectorization on 8 processors) of 2.7. Gigaflops and costs between $20m and $30m. The Intel iPSC/860 has 128 ...
  26. [26]
    Cray EL - Cray-History.net
    The Cray-YMP EL range of computers were manufactured from 1992 until 1996. A direct development of the XMS project this range included the smallest and ...
  27. [27]
    The Cray Y-MP EL: An overview - saint-christophers-hosting.com
    Jan 26, 2021 · Launched in '89, Supertek managed to sell about ten “Supertek S-1” machines and then started development of a Cray YMP processor implementation.Missing: background | Show results with:background
  28. [28]
    [PDF] CRAY Y-MP EL Functional Description HR-04027
    The operating registers and functional units of each computation section are associated with three types of processing: address, scalar, and vector. Address ...Missing: choke | Show results with:choke
  29. [29]
    Cray Inc. - Company Profile, Information, Business Description ...
    Within one month in late 1991 , Cray introduced an entry-level system priced at about $340,000 called the Y-MP EL and its fastest vector supercomputer to date, ...<|control11|><|separator|>
  30. [30]
    [PDF] Performance of Various Computers Using Standard Linear ...
    Nov 10, 2001 · The first results listed in Table 1 involved no hand optimization of the LINPACK benchmark. ... Cray Y-MP/832 (8 proc. 6 ns). CF77 4.0 -Zp -Wd-e68.
  31. [31]
    [PDF] The CRAY Y-MP8 Supercomputer Systems
    With up to eight proces- sors, the CRAY Y-MP8E system offers the largest central memory, SSD, and I/O capacity ever available on CRAY. Y-MP supercomputers. The ...
  32. [32]
    Cray Y-MP - Wikipedia
    The Cray Y-MP was a supercomputer sold by Cray Research from 1988, and the successor to the company's X-MP. The Y-MP retained software compatibility with ...
  33. [33]
    CRAY X-MP Super Computer (C/N) - Jurassic-Pedia
    Each CPU had a theoretical peak performance of 200 MFLOPS, for a peak system performance of 400 MFLOPS. ... In Jurassic Park, three Cray X-MP computers were used ...
  34. [34]
    Performance Attributes for Code and Workload Analysis On Cray X ...
    Although the clock pe riod on the Y-MP was reduced by 29% relative to the X-MP, increases of 52% were observed in instruction fetch rates and vector MFLOPS, ...Missing: predecessors | Show results with:predecessors
  35. [35]
    FAQ-5 Cray Machine specifications
    This FAQ lists specifications of various Cray supercomputers, including models like Cray 1s, Cray 2, and T3d, with varying memory and performance.
  36. [36]
    VP2600/10 - TOP500
    5.00 GFlop/s. Software. Operating System: Unix. Ranking. List, Rank, System, Vendor, Total Cores, Rmax (GFlop/s), Rpeak (GFlop/s), Power (kW). 06/1996, 395 ...
  37. [37]
    [PDF] MASTER - OSTI.GOV
    A PERFORMANCE COMPARISON OF THREE SUPERCOMPUTERS: FUJITSU VP-2600, NEC SX-3, AND CRAY Y-MP. AUTHOR(S):. SUBMITTED TO: Margaret L. Simmons. Harvey J. Wasserman.
  38. [38]
    A performance comparison of three supercomputers
    Aug 1, 1991 · A performance comparison of three supercomputers: Fujitsu VP-2600, NEC SX-3, and CRAY Y-MP.
  39. [39]
    A performance comparison of four supercomputers
    For compari- son purposes, we include results obtained using a CRAY Y-MP and a. CRAY Y-MP C90. The Y-MP was introduced by Cray Research, Inc. (CRI) in 1988 ...<|control11|><|separator|>
  40. [40]
    June 1993 - TOP500
    The top 3 supercomputers in June 1993 were CM-5/1024, CM-5/544, and CM-5/512. The CM-5/1024 had 1024 cores.Missing: precursor 1988-1993
  41. [41]
    How much would it cost you to build a supercomputer? I bet ... - Quora
    Dec 25, 2017 · The Cray Y-MP. It would cost $13,000,000 today. This is the Cray Y-MP. It featured up to 8 liquid cooled processors ...
  42. [42]
    $$110 Million for New Supercomputer Sought by Center on UCSD ...
    Jul 23, 1987 · The $110-million NSF request, which was submitted about a month ago, would allow the center to acquire a $40-million Cray YMP. That model is ...
  43. [43]
    [PDF] Early Prediction of MPP Performance: The SP2, T3D, and Paragon ...
    This approach has be used in the. NAS benchmark[23], where the flop count is determined by an execution run on a Cray Y-MP. Throughout the rest ...
  44. [44]
    [PDF] Why Multiprocessors? Contact and Communication
    Shared-memory MP design. 6. Shared-memory MP design ... Even Crays became parallel: X-MP (2-4) Y-MP (8) ... • Can support data parallel model on SM or MP.
  45. [45]
    How large was the Cray Y-MP, and how much power did it need?
    Sep 3, 2019 · How large was the Cray Y-MP, and how much power did it need? Hmm. I never actually measured it, and it also depends on which version (the ...
  46. [46]
    Cray-2 - Wikipedia
    The power consumption of the Cray-2 was 150–200 kW. Research conducted at the Lawrence Livermore National Laboratory in the early 1990s indicated that to a ...
  47. [47]
    [PDF] NASA's Supercomputing Experience
    flows over wing bodies on the Cray XMP/4. Today the Cray YMP/8 enables simulation about very complex geometries such as the space shuttle configuration.Missing: specifications | Show results with:specifications<|separator|>
  48. [48]
    [PDF] Automatic Parallelization: - A Comparison of CRAY fpp and KAI KAP ...
    their ability to vectorize and parallelize 25 scientific benchmarks for a CRAY Y-MP ... Memory access time:17 clock periods (107 ns). Memory bank cycle time ...
  49. [49]
    Model SUNYA/NCAR GENESIS1.5 (T31 L18) 1994 - PCMDI
    Apr 19, 1996 · The AMIP simulation was run on a Cray Y/MP computer using a single processor in a UNICOS environment. Computational Performance. For the AMIP ...<|separator|>
  50. [50]
    A History of LLNL Computers
    CRAY-2 supercomputer featured four CPUs and 67 million 64-bit words of memory. With delivery of this machine, the Lab retired its CDC 7600. ... X/MP & Y/MP. 1985.
  51. [51]
    An analysis of the indirect boundary element method for seismic ...
    Significant savings can be accomplished in the BEM by the use of alternative techniques to the Gaussian elimination. ... CRAY-YMP supercomputer. Fig.8 ...
  52. [52]
    Ultrahigh-performance FFTs for the CRAY-2 and CRAY Y-MP ...
    In this paper a set of techniques for improving the performance of the fast Fourier transform (FFT) algorithm on modern vector-oriented supercomputers is ...
  53. [53]
    Application of the Ewald method to cosmological N-body simulations
    The present simulations were carried out on the CRAY-YMP at the Pittsburgh Supercomputing Center and on the Hitac 5-820/80 at KEK (National Laboratory for High ...
  54. [54]
    [PDF] Chemical calculations on Cray computers
    The CRAY. Y-MP is a natural successor to the X-MP, with a clock period of 6 ns and up to 32 MW of memory. Its performance characteristics are very similar to ...
  55. [55]
    Performance Attributes for Code and Workload Analysis On Cray X ...
    Although the clock pe riod on the Y-MP was reduced by 29% relative to the X-MP, increases of 52% were observed in instruction fetch rates and vector MFLOPS, ...Missing: datasets | Show results with:datasets
  56. [56]
    Cray Y-MP - NASA Technical Reports Server (NTRS)
    Nov 1, 1988 · This video shows the installation of the Cray Y-MP, a computer four times faster than any other computer at Ames.Missing: Los Alamos Pittsburgh UK Meteorological University Texas total
  57. [57]
    Computing on the mesa | Los Alamos National Laboratory
    Dec 1, 2020 · The 1982 Cray X-MP possessed two central processors, while its successor, the Y-MP, featured up to eight. These machines formed the bulk of ...
  58. [58]
    [PDF] Delivering Insight Final - Advanced Simulation and Computing
    DP is responsible for managing the entire nuclear weapons complex. A second component of the complex comprises nuclear weapons design and engineering,.
  59. [59]
    CRI Cray Y-MP2 - Castle | Computational and Information Systems ...
    The Cray Y-MP2 'Castle' was a supercomputer used for climate simulation, named after Castle Peak, and was the first in North America dedicated to this purpose.
  60. [60]
    [PDF] TOP500 Supercomputer Sites - The Netlib
    Nov 18, 1996 · Y-MP T94/4128. Muenchen Germany /1996. 7200 . 393. Cray. Los Alamos National Laboratory. Research. 4. 5735 . Y-MP T94/4128. Los Alamos USA /1995.
  61. [61]
    Decommissioned PSC Systems
    The CRAY T3D system was the first in a series of massively parallel processing (MPP) systems from CRAY Research. T3D's are tightly coupled to CRAY Y-MP and C90 ...
  62. [62]
    CRAY RESEARCH EXPLAINS WHAT IT HAS FOR 1990s, WHY ...
    Nov 27, 1991 · So Cray is confident that its Y-MP C90 (C90 is an acronym for Cray for the '90s), is a major breakthrough, a true supercomputer that is here ...<|control11|><|separator|>
  63. [63]
    Cray Machine Families up to year 2000 - Cray-History.net
    The Cray-YMP EL range of computers were manufactured from 1992 until 1996. The range included the smallest and cheapest system ever delivered to customers. The ...
  64. [64]
    SGI & CRAY: WILL A MARRIAGE OF NECESSITY END HAPPILY?
    Mar 1, 1996 · As it is, the leaders of SGI and their new colleagues from Cray have a reasonably good chance to make this merger an exception that really ...Missing: Y- | Show results with:Y-
  65. [65]
    Cray Research Shows Path To Unified High-Performance Architecture
    Nov 14, 2024 · "Our merger with Silicon Graphics enables Cray Research to focus completely and intently on high- end supercomputing while the combined ...<|separator|>
  66. [66]
    [PDF] UNICOS/mk Status and Update - Cray User Group
    ABSTRACT: The UNICOS/mk operating system has been introduced recently on the CRAY. T3E mainframe. This operating system is a modularized version of the ...
  67. [67]
    Cray Inc. Company (Unicos) - Operating-system.org
    Dec 22, 2023 · The Cray Y-MP system has reached a total performance of up to 2.3 GFLOP together with several CPUs and each 333 MFLOP in 1988. ... - UNICOS ...
  68. [68]
    A Supercomputing Star--And Why It Fell | InformationWeek
    Sales peaked in 1994 at $922 million. But by then, Cray had become less of a factor in setting technology directions. Steve Oberlin, Cray's former chief ...
  69. [69]
    Sneakers (1992) - Trivia - IMDb
    ... Cray Y-MP seen in the movie. He is also often referred to as the "Father of Information Security" by seasoned veterans of the computing industry. Helpful•126. 2.
  70. [70]
  71. [71]
    Seymour Cray, the father of supercomputers and vector processing
    Dec 5, 2015 · Cray became practically synonymous with supercomputing for decades, in fact Cray is widely known as the father of supercomputing. Today we ...
  72. [72]
    Cray Y-MP EL - X2101.2001 - CHM - Computer History Museum
    Cray Y-MP EL. Cray Y-MP EL. Cray Y-MP EL - Image 1. Item Details. Date: 1992 (Made); Type: Physical Object; Catalogue number: X2101.2001; Other identifying ...